Tokens today are mostly a way for teams to double-dip on equity: the market prices them like a claim on the business from the day they list, and the team owes that assumption nothing. @umia_finance is an upcoming EVM venture-creation platform built to close exactly that gap — make the token legally own the venture (the IP, the revenue, the treasury) and everything else follows..

The double dip

Priced like equity from day one

The market does not wait to be told what a token is. The moment a project token lists it trades on the business: revenue milestones move it, team departures move it, a competitor’s raise moves it. Buyers underwrite the token as a claim on the venture immediately and by default, because that is the only story under which most project tokens are worth anything at all.

Owed like nothing

The team sells into that assumption without inheriting any of its obligations. Nothing routes revenue to the token. Nothing assigns the IP. The treasury sits in a multisig the team controls, and the token’s “governance” binds nothing off-chain: a team that ignores its DAO faces a Discord thread, not a court. The standard playbook — list at a high FDV on a thin float, let the unlock schedule sell the assumption down for two years — is the double dip made mechanical: raise like equity, owe nothing. When the link helps, it’s in every pitch; when it binds, the token is suddenly a utility token. None of this is an edge case. It’s the equilibrium the structure invites, and it’s paid for by holders.

The next step

If the market insists on pricing tokens as equity, the durable fix is to make them equity in fact: the token legally owns the venture’s assets, the treasury sits outside the team’s reach, and governance produces decisions that bind the legal entity. That bundle is the actual next step for tokens. Two structures now sell it; Umia is the newer one.

What Umia does

Every launch on Umia gets four components wired into one loop: an auction funds a non-custodial treasury, the treasury pays a programmatic budget, decision markets govern everything above the budget, and a legal wrapper binds the off-chain company to what the markets decide. Umia’s name for the result is a liquid venture: the token owns the revenue, IP, and treasury from day one, so the TGE is the starting line rather than the peak.

The auction: price discovery during the raise

Launches run through a Continuous Clearing Auction built on Uniswap v4. Tokens release on a fixed per-block schedule; each block, incoming bids clear against that block’s release, and the clearing price ticks up whenever demand exceeds it: price discovery happens during the raise, monotonically, from floor to final clearing. Bidder access is gated by zkTLS proofs (via Reclaim): a project can verify who its bidders are (geography, holdings, community membership) without collecting their data. Proceeds land in the treasury and the token trades immediately on Uniswap v4 with protocol-owned liquidity.

The treasury: a budget the team can’t override

Non-custodial: the team cannot spend it. It auto-pays a fixed monthly budget by contract, $120K/month in Umia’s own case. Anything beyond the budget (a one-time spend, a mint or burn, an investment, a liquidation, a change to the budget cap itself) needs the markets. Venture revenue flows back into the same treasury, so the loop closes.

The decision markets: futarchy as the board

Board-level decisions settle by futarchy: a decision lists any number of options, conditional markets price the token under each, and the highest-priced branch executes. Approve/reject is just the two-option case. Founders run operations; markets settle the board votes: treasury actions on-chain, and through the wrapper, IP, contracts, team changes, pivots, and dissolution off-chain.

The wrapper: markets that bind a legal entity

Each venture is a Cayman entity built on MetaLex’s BORG framework (Umia itself is a segregated portfolio of Umia Launcher SPC), with vesting in MetaVesT contracts. The wrapper turns “the markets decided” into an instruction the legal entity must execute — the piece that converts the double-dip diagnosis into an enforceable claim. No such structure has yet been tested in a dispute.

The $UMIA token

Umia is the first venture on its own platform, and all protocol value routes to $UMIA through a fee switch: ~1% of decision-market volume across the platform, ~0.5% of spot swaps on every launched token, plus a slice of each launched project’s supply into the treasury.

The sharpest line in the tokenomics is the team side. Team-attributed tokens total 15M (30% of supply), but only 5M vest on schedule; the other 10M unlock only at 30-day TWAP multiples of the TGE price (1M at 2×, 2M at 5×, 2M at 10×, 3M at 20×, 2M at 50×). If the price never performs, two-thirds of the team’s upside never exists. That is a real structural answer to the low-float-high-FDV complaint, though price-triggered unlocks also hand the team a direct interest in engineering price around trigger windows; 30 days of TWAP makes that expensive, not impossible.

Who is behind it

The team is public and the resume checks out. CEO Francesco Mosterts came from Point72 and founded Chainbound, the Ethereum R&D lab that incubates Umia; Nicolas Racchi is CTO. The cap table has named checks (cyber.Fund, Maven 11, Robot Ventures, Delphi Ventures, Bankless Ventures) plus angels including Hasu and Anthony Sassano, and the contracts carry a Certora audit. All of that is verifiable.

The MetaDAO question

The overlap

Umia positions itself against four categories (memecoin launchpads, ICO platforms, curated launchpads, equity VC) and sweeps its own comparison grid. The category it omits is the one it belongs to. @MetaDAOProject has run this design on Solana since its first ICO opened on April 2, 2025 (mtnCapital, which closed at $5.76M): 20 successful raises totaling ~$44.9M, headlined by Solomon and Ranger at $8.0M each, plus 60+ failed micro-ICOs since launches went permissionless in March 2026, and two raises live as this is written. On most vectors, Umia is MetaDAO re-implemented:

The genuine differences

1. Price discovery during the raise. MetaDAO runs a fixed-price 4-day ICO: everyone pays the same FDV, and discovery waits for listing. Umia’s CCA discovers price inside the sale itself, block by block. This is Umia’s strongest mechanical claim.

2. Chain and distribution. Ethereum and Uniswap v4 versus Solana and MetaDAO’s own futarchy programs. Less a feature than a market: Umia brings the model to capital that lives on Ethereum.

3. Fee placement. MetaDAO taxes secondary trading: 50bps on ownership-coin AMM volume. Umia taxes the governance mechanism itself: ~1% on every decision-market trade, plus 0.5% on spot swaps.

4. Disclosure maturity. MetaDAO publishes its settlement guardrails (3-day markets, a 24-hour pricing delay, ±3% pass thresholds, movement-capped TWAPs) because thin conditional markets can be moved by a handful of traders — the median MetaDAO decision draws 13. Umia has not published its equivalents.

5. Multiplex decisions. MetaDAO’s futarchy is binary: every proposal is a pass/fail pair of conditional markets against the status quo. Umia’s decision markets take any number of options in a single decision and execute the highest-priced one, so a question with five candidate answers settles in one market instead of five sequential proposals.

My thoughts

If you’re evaluating this raise, the comp is MetaDAO’s live number, and the number comes apart into two pieces. META trades at a $140M FDV. Underneath it sits an ecosystem Umia doesn’t have: roughly $80M of launched-token market cap, about $22M of treasury value across its ventures, and $800K–$1.6M a year of revenue at the current run rate. At 10× revenue, that existing ecosystem is worth $8M–$16M of the FDV. The other $124M–$132M is the market’s price on MetaDAO’s execution: the dealflow, and a team that earned its guardrails mid-flight (the settlement rules, the fee model, the lessons of the permissionless flood were pivots, not the plan).

Umia deserves none of the first piece and a fraction of the second, and the fraction is the whole question. I can’t measure this team’s shipping ability, and neither can anyone outside it; what I can measure is how hard the game is for the incumbent. Sixteen months of MetaDAO dealflow produced 20 successful raises, two of them at $8M, against 60+ failed micro-ICOs: good deals are scarce even for the team that invented the category and learned its whys by pivoting through it. Umia’s team arrives with credentials and none of that earned understanding, and in a game this new, credentials are not evidence. That earns a heavy discount until the team is seen in action. I’d hold the fraction at 15–33% of MetaDAO’s execution value, which puts Umia’s fair start between roughly $19M and $44M FDV, and the first handful of launches turns it from a guess into a measurement.

So the way I’d play it is to treat the band as the bid. Wherever your conviction sits inside $19M–$44M is the most you pay; above it you pass, and entry below it is margin. Then let the first few raises grade the estimate — the quality of the ventures the team lands and the decisions it makes around them — and add or cut as the fraction resolves. The underwrite is $UMIA closing most or all of the distance to META’s valuation: a 3–8× from inside the band even if META stands still.

@flint_trade_ went live with a pitch aimed at a specific absence: the deepest market-making balance sheets in crypto quote on every major CEX and mostly skip Solana spot. Its answer is to run the exchange for them: one shared on-chain venue where a maker connects, sends prices, and never touches a Solana program.

There are three ways to make markets on Solana spot. Build your own prop AMM: the full stack, months of building with an engineering team that can win a compute-unit war, and you keep the entire spread. Quote on Flint: a bot you can write in weeks, a 0.01–1 bp venue fee, and execution rails identical to every other maker’s. Or supply a public DEX: zero integration, passive terms, and the adverse selection priced in. Flint wins for the desk that already has a pricing model and doesn’t have a Solana team, the CEX or cross-chain market maker for whom the build is unjustifiable and the public pool is uninvestable. For everyone else, one of the other two routes is still the better trade.

How Flint works

flint is one shared on-chain venue that multiple independent market makers quote into. A maker deposits inventory with Flint’s program and streams prices against it; takers arrive through aggregator routing (OKX DEX and Titan at launch) and trade against the best prices shown. The sequence in between works like this:

1. A maker registers a quoting authority (a hot key that signs quotes but not withdrawals), deposits inventory, and picks a quoting style: an offset around an oracle price, an explicit order list, or a linear distribution. The maker’s remaining problem is producing a fair price.

2. The maker’s bot streams that fair price. Flint’s client library packs it into an optimized transaction, signs it, and lands it; priority fees, compute budget, and sequencing are the venue’s problem. A refresh sends only the new price, and the program shifts the maker’s whole book around it without rewriting order state. This is the lightweight-update trick the prop AMMs engineer in-house, sold here as a library call.

3. The program merges every maker’s intents into one virtual orderbook. Nobody owns the book; at each price level, fills allocate pro-rata across the makers quoting there. A tighter quote buys priority, not exclusivity, and every maker’s quotes are visible to the rest of the venue.

4. A taker’s swap crosses the merged book, and each filled maker gets a receipt: confirmed, failed, or unknown. Inventory risk, skew, and the decision to pull quotes stay with the maker; the venue only guarantees that a sent quote reaches the chain.

What the maker keeps is exactly what Flint cannot commoditize: the fair-value model, the capital, and the risk discipline. What the maker gives up is the execution layer as a weapon.

Does the shared infrastructure keep up?

The mechanism only matters if the rented rails are competitive. The measurable half of that question is the cost of a quote update: Jito orders transactions by tip per compute unit, so the fewer CUs an update burns, the cheaper it is to hold queue priority — this is the number the prop AMMs spend engineering careers driving down. I sampled quote-update transactions for every live prop AMM with a measurable quote flood over one ten-minute window (August 23, 2026, 12:00–12:10 UTC) and took the median compute units per update transaction.

Flint’s update lands at 281 CUs — the cheapest live quote update measured, under BisonFi’s 404, Aquifer’s 486, and Quantum’s 638, and beaten only by HumidiFi’s reported 143 from 2025. The fleet Flint has to keep up with is the one on the chart. Three venues are off it by design: GoonFi exposes no visible update mechanism at all, Ghost runs a sparse crank instead of a flood, and Deriverse maintains a visible order ladder at ~153,000 CUs — a different design, not a quote push.

Three caveats keep this honest. Archer and AlphaQ batch up to ten and sixteen markets into one transaction, so their per-market cost is far below their per-transaction bar — AlphaQ’s 2,068 CUs over sixteen pools is roughly 130 per market, plausibly the cheapest per quote on the chart. Compute units are also only the denominator of queue priority; a venue paying larger tips lands ahead of a leaner program, and tips aren’t measured here. And unit cost isn’t cadence: Quantum fired 28,700 updates in the sample’s ten minutes while venues a third its cost updated far less. Cheap and stale loses to expensive and fresh.

On the axis a maker can’t control — what an update costs on the shared rails — Flint currently holds up against every venue running its own stack. Whether it stays there is checkable the same way: the day Flint’s update cost drifts above the fleet’s, that case fails.

What each route costs

Building your own prop AMM

The observed bill, from the shops that paid it: a Jump-caliber lead (HumidiFi’s Kevin spent nine years there), a systems team or an infrastructure partner (Temporal, in HumidiFi’s case), and a CU-optimization war. HumidiFi ground its oracle update down for six months — from ~300 CUs to the 143 measured in 2025, with the team claiming sub-50 by the end — because the update that lands first wins the quote. Add bare-metal nodes at $2,200–2,900 a month each across regions, ShredStream, SWQoS, leader-proximity engineering, and Jupiter onboarding: an audit, an SDK against the jupiter-amm-interface crate, and quote-parity tests before routing turns on. The reliability bar is part of the price: HumidiFi’s oracle sender crashed in the October 10, 2025 volatility spike and the halt cost over $500,000.

The payoff side is why anyone pays it. The model runs $5–10M of vault inventory against $500M–$1B+ of daily volume (Step Data measured $5.33M of pool TVL against $819M of trades in a single day), and the operator keeps the full oracle-to-execution spread, plus every microstructure advantage it can engineer. HumidiFi did it in about four months: first trade in June 2025, largest DEX on Solana by October.

Quoting on Flint

Everything a maker has to build is a bot. It reads a fair price from an external source, computes one two-sided quote with a spread and an inventory skew, sends it, and pulls it when the price feed goes stale or a risk limit trips. Flint’s own maker-bot spec assumes a quote refresh every few seconds and a single price level per side; one engineer can write it in weeks. The Solana-specific costs of building your own venue disappear entirely: no on-chain program to write, no transaction-landing infrastructure, no compute-unit race, no Jupiter onboarding process.

Two new costs take their place. The first is the fee Flint charges on each trade, set per pair and readable from its public API: 0.01 bp on stable-to-stable pairs, 0.5 bp on majors like WSOL/USDC, and 1 bp on volatile pairs (pulled from the mainnet catalog, August 24, 2026). For comparison, public pool fee tiers on the same pairs run 1–30 bp, so the fee is small next to the spread a maker quotes. The second is shared infrastructure: every maker lands through the same rails, so the pricing model is the only remaining edge.

Supplying a public DEX

Permissionless and instant: deposit into an Orca or Raydium pool, or quote a public CLOB like Phoenix, with no venue’s permission and no SDK. The costs are structural. A passive LP position is a standing offer at a stale price: it owes the router a fill on demand and collects a pool fee for it, which is precisely the position the prop AMMs were built to trade against. Active quoting on a public venue means paying full transaction costs on every update with none of the prop-AMM machinery (no few-hundred-CU refresh, no cancel priority), so the operator eats the latency race at retail prices. Public pools are where passive capital gets paid to take the other side, not where a desk runs a spread business.

Who each route is for

Where does the edge live?

A desk that wins by landing first should own the stack. Flint’s shared rails neutralize the thing it is best at, and the venue-scale outcome (HumidiFi’s DEX-#1 run, a token) only exists on your own program.

A desk with a pricing edge and no Solana stack (a CEX desk, an other-chain prop shop, a quant team with a model and no infra) is Flint’s segment, and its founders’ résumés (Jito Labs, Jump, dYdX Foundation) say they scoped it deliberately. The full-stack build was the reason these desks skipped Solana; a bot is not.

Capital without a model belongs on public LP terms, which are the honest product: fee income for warehousing adverse selection, no ops. Flint adds operational burden with no model to justify it.

Whether quoting on Flint is worth a desk’s time comes down to three conditions. The desk does not want to enter the engineering competition that running a prop AMM has become. It believes the volume moving through Solana spot aggregators — the flow that lets prop AMMs turn $5–10M of inventory into $500M–$1B of daily trades — is worth going after. And Flint’s infrastructure stays competitive with the vertically integrated venues its makers quote against. The first two are questions a desk can answer for itself today. The third is measurable on-chain: whether Flint’s book keeps winning fills against the incumbent prop AMMs through 2027.

Thesis

For most of its history, crypto has traded claims on itself: the top assets were tokens of protocols building crypto infrastructure, and very few represented businesses, objects or value that lived off-chain. Where crypto projects will actually find product-market fit is where the technology enables price discovery on an asset class in ways its existing market structure cannot — by being cheaper, faster, less gatekept — or simply open when the incumbent isn’t. And the real money is in hosting that discovery, not just the trading: venues where the price is actually made can charge far higher fees and take rates on the volume that passes through them, and the discovery itself attracts more volume. Finding the platforms where off-chain assets and value are being priced on-chain in ways they couldn’t be before, and getting exposure to them early, will be one of the key ways to earn outsized returns in crypto over the coming years. The four case studies below are four different failures of the incumbent structure: it closes (Hyperliquid), it doesn’t exist (PreStocks), it gates who can even create the asset (MetaDAO), and it overcharges (Collector Crypt).

Case studies

Hyperliquid (off-hours trading)

@HyperliquidX

is the simplest version of the thesis, because the way the existing market structure fails here is total: it closes. CME’s commodity session is closed for 49 hours every weekend and US equity venues are closed 65 of every 168 hours, while Hyperliquid trades continuously — so for those windows the on-chain venue is not competing on cheaper or less gatekept, it is the only place the asset can be priced at all. Since HIP-3 launched in October 2025, anyone staking 500K HYPE can deploy a perp market on any asset with a price feed (builder markets covered 250+ US stocks and ETFs by March 2026), and traditional financial assets have since grown to more than 30% of platform volume: price discovery happening on-chain during hours when it previously happened nowhere. And the volume is worth more to Hyperliquid than it would be to the incumbent: its base taker fee is 4.5 basis points of notional, while CME’s exchange fee on a 1,000-barrel crude contract is about $1.50 — a fifth of a basis point at $70 oil. That is roughly a 20× gap per dollar traded, so when the calendar hands Hyperliquid flow, it prints.

The first clear case was silver. During the squeeze in late January 2026, when silver traded through $110/oz and gold through $5,000, the Hyperliquid silver perp averaged $1.25B in daily volume and reached $4.09B on February 5, 68% of all builder-market volume that day. COMEX was closed on weekends throughout the squeeze, which left the perp as the only continuously traded silver market during those windows.

Oil followed a month later, with the catalyst itself arriving on a weekend. US and Israeli strikes on Iran began in the early hours of Saturday, February 28, while every traditional commodity venue was closed. The oil perp repriced from $68.72 to above $70.60 against a Friday close of $67.28, and CME reopened at $75. Weekend oil notional grew from roughly $25M to $550M over three weekends. TD Securities estimated that about 80% of the WTI move was priced on Hyperliquid before CME reopened, and by the fourth weekend the difference between the perp’s pre-open price and the CME reopen was 4.9 basis points.

The third case is Korean equities. In July 2026 the KOSPI posted its strongest monthly gain since 2001, then fell 11% in a single session on July 29. The SK Hynix perp moved an hour before Seoul opened that morning, falling 19% and liquidating $57.4M in positions after an anomalous print on a Korean alternative venue; the underlying stock closed the session down 15%. The episode showed both properties of the market: off-hours Korean risk now reprices on Hyperliquid before the home exchange opens, and that pricing depends on oracle feeds a single bad print can move.

The caveats cluster in three places. Weekend discovery is narrower than the headline episodes suggest: in the same 0xArchive sample that documented the oil weekends, the broad equity tail’s weekend price beat Friday’s close barely more than half the time, and weekend depth ran at about a third of weekday levels — the effect concentrates in the handful of markets where weekend volume actually shows up. The pricing machinery is only as reliable as its oracles, and the SK Hynix episode put $57.4M of liquidations on a single anomalous print from a thin Korean venue. And the moat is the incumbents’ calendar rather than anything Hyperliquid controls. CME already moved: its 24/7 crypto futures went live on May 29, 2026, and traded about $50M of notional over their first weekend — a tenth of a single Hyperliquid oil weekend, but the test of the calendar moat has started. If round-the-clock hours reach metals, energy, and equities, the off-hours window, currently 39% of the week for US stocks, stops being uncontested.

PreStocks (private companies)

@PreStocks is the less-gatekept version of the thesis. A private company’s shares have no trading venue: price is set in funding rounds and tenders, months apart, behind accreditation rules and transfer restrictions. That makes private markets the largest asset pool in the world with no continuous price — a class that just produced history’s largest IPO (SpaceX, $1.75T, June 2026) after a private life spent without a live quote.

PreStocks buys exposure through SPVs and issues Solana tokens against it: eight names (SpaceX, OpenAI, Anthropic, xAI, Anduril, Neuralink, Polymarket, Kalshi), trading 24/7 with no minimums and no accreditation gate. For the still-private names, the token chart is the only continuous price that exists anywhere, and the demand is proven: before SpaceX listed, five separate issuers had tokenized it. When Anthropic and OpenAI disputed the structure in May 2026, the tokens repriced. As of August 5 the SpaceX token trades at $92.28 against a $109.29 Nasdaq price and Polymarket at $141.60 against $163.97: a 14–16% discount that is the market’s live estimate of legal-enforceability risk. No private-market structure produces that number. The venue economics are the mirror of it: with no reference price to arbitrage against, PreStocks market-makes its own tokens at whatever spread the market bears.

The problems cluster in two places. The business is small: about $3.2M of weekly volume across all eight tokens, $83K of SpaceX pool liquidity, floats of 1,300–10,200 tokens. And the claim is weak: after the May dispute the tokens fell 30–40%, four markets were unwound outright (Kraken, Perplexity, Discord, Epic Games), and every mint carries a permanent-delegate flag that lets the issuer claw tokens back from any wallet — holders own a revocable exposure, not an asset. The wedge also expires name by name: when SpaceX listed, its discovery moved to Nasdaq and the token became a tracker. And the corner is not automatically lucrative: Ventuals, which ran pre-IPO perps on Hyperliquid with no shares behind them at all, wound down its private-company markets in 2026. Whoever ships a structure that survives issuers’ transfer restrictions inherits demand five tokenizers already proved; until then, the 15% discount is the price of the gap.

MetaDAO (liquid venture)

@MetaDAOProject is the version of the thesis where the asset is born on-chain instead of wrapped onto it. Early-stage venture is the most gatekept market there is: priced by a handful of investors, round by round, untradeable in between. And where PreStocks’ claim on its underlying is disputed, MetaDAO removes the problem by construction — the company raises, holds its treasury, and governs on-chain from day one, so the token is not exposure to ownership; it is the ownership.

The mechanics: an ICO seeds an on-chain treasury (drawable only against a monthly spending limit) and a spot AMM where the ownership token trades continuously; board-level decisions settle in futarchy markets that price the token under pass and under fail, with the winning TWAP executing. Since April 2025 the platform has completed 17 raises totaling about $44.9M, the largest at $8.0M (Solomon and Ranger), and twelve live DAOs spend roughly $858K a month from auditable treasuries. MetaDAO takes 50 basis points on ownership-coin volume, roughly ten times what DEX pools sustain on assets that trade anywhere else. The fee holds because there is no anywhere else: the futarchy AMM is the only venue where the ownership coins trade.

The problems are thinness and scale. The median futarchy decision draws 13 traders, which is why the guardrails exist (three-day markets, a 24-hour pricing delay, ±3% thresholds, movement-capped TWAPs): a 13-trader market is a real price only until someone has a reason to bend it. Since launches went permissionless in March 2026, 60+ micro-ICOs have died below their minimum raise. And the venue is boutique: $44.9M over sixteen months is one mid-sized seed fund. The bet is that the structure compounds to the next thousand companies rather than the current twelve.

Collector Crypt (collectibles)

@Collector_Crypt is the cheaper-and-faster version of the thesis. Trading cards have an incumbent venue with real prices; it just charges double-digit fees and takes a week to settle a sale. Collector Crypt, Courtyard, and a handful of other platforms attack it the same way: the physical card is authenticated and graded once, then locked in a vault and issued as a token on-chain. From that point the title moves instead of the object, and the card trades like any other digital asset until someone burns the token to have the physical shipped out.

The friction that removes is most of what card trading currently is. A graded card on eBay prices once per sale, in an auction that runs about a week, with a 13.25% marketplace fee, shipping both directions, and the risk of damage or fakes on every resale: the authentication work is redone every time the card changes hands. Vaulted, the same card is authenticated exactly once and then settles in seconds for cents, and the buyer of the token never takes shipping or counterfeit risk at all. The incumbent structure prices a card at double-digit cost per transaction; Collector Crypt charges 2% (a 1% platform fee plus a 1% royalty) — a sixth of eBay’s cut, and still a take rate most crypto venues never see.

Nothing in that model is specific to Pokémon. It generalizes to any collectible that can be authenticated and vaulted (Collector Crypt’s own treasury already spans Pokémon, One Piece, sports cards, Magic, and Yugioh), and the same vault-and-token structure maps onto watches, sneakers, coins, and comics without modification. Wherever a physical asset’s trading cost is dominated by re-authentication and shipping rather than by the asset itself, the tokenized version wins on friction alone.

The caveat is what the revenue mix says about demand. The product that brought the inventory on-chain is not the marketplace but the Gachapon: randomized packs of vaulted, tokenized cards at $50, $250, and $1,000+ price points. Gambling built the venue. People do not show up to tokenize cards, they show up to pull packs, and feeding the packs is what filled the vault with tens of millions of dollars of physical inventory. But the same mix measures the distance left: the marketplace itself, the continuous secondary market the thesis actually cares about, generated under 2% of the revenue. Demand for holding and trading tokenized cards on their own merits is still mostly untested — today the venue exists because a casino needed inventory, not because collectors moved their trading on-chain. The test for the thesis is whether secondary volume grows into the vault the gacha built.

Final Thoughts

Only Hyperliquid is clearly hosting price discovery at scale today: 80% of a WTI move priced before CME opened, Monday opens importing its weekend moves. The others haven’t proven it yet. Collector Crypt’s discovery rides a casino (secondary trading is under 2% of revenue), MetaDAO’s decision markets draw a median of 13 traders, and PreStocks’ quotes sit on a disputed legal claim. One proof, three early attempts.

So the bet is on the demand for prices, not on the tickers. Every case found assets that real money wants priced and the existing structure leaves unpriced: on weekends, before the IPO, between funding rounds, between auctions. Someone will get paid to make those prices; whether it is these four platforms or their successors is open.

The filter is discovery itself. A venue is worth underwriting only if other venues reprice off its numbers, because that is what supports a take rate above the basis points earned by venues that import prices. All four already charge that way: fees roughly 20× CME’s per dollar of notional, 50 basis points where DEXs sustain single digits, 2% where marketplaces compete in fractions, and a spread set with no reference price to undercut it. Applied forward, the lens points where this paper already brushed: every other exchange’s closed hours, the private names still without a surviving structure, and the vaultable collectibles still priced auction by auction. If by 2028 no venue in these categories holds that premium, the thesis fails. My read: the unpriced assets get priced, and the tickers are the open question.

Thesis

@omnipair bundles two products that are each already losers on their own — full-range passive market making, which no unsubsidized capital does as an investment anymore, and utilization-curve lending, whose rate can’t price collateral risk on exactly the assets it exists to list. The bundle fixes neither; it stacks the same microcap exposure twice and gates the exit in the one state you’d use it — while Solana’s market structure specializes into venues that do each job properly.

What Omnipair is

Omnipair is a @MetaDAOProject-launched “GAMM”: each pool is simultaneously a UniV2-style constant-product AMM (full-range x·y=k, static fee, typically 25bps) and an isolated lending market. LPs deposit both sides; that same liquidity is the loan book. Borrowers post one side as collateral to borrow the other; leverage is recursive borrow-and-swap inside the pool. There are no external oracles — all risk math runs on a double-EMA of the pool’s own price, with impact-aware collateral valuation and liquidation by debt write-off: seized collateral is absorbed into pool reserves and, in the docs’ own words, “bad debt is socialized to LPs through reduced pool reserves.” LP withdrawals carry a 1% fee and are gated by utilization when liquidity is lent out.

The pitch: most Solana volume is long-tail, long-tail assets can’t get listed on oracle-based money markets, and a pool that is its own oracle can list anything. The question is never whether the product can exist. It’s whether anyone providing the liquidity can survive providing it. Five months post-launch the market has given a preliminary answer — roughly $550k of TVL, $46k of borrows, and no new pool created since late May. The reflex is to call that early. But early implies a growth curve, and the rest of this paper argues there isn’t one: these numbers are the steady-state output of the position the design manufactures. The LP’s problem is the composition of the flow, not the quantity — and composition doesn’t improve with scale.

The LP leg: the worst seat at the table

In a router meta, a passive pool is a toxic-flow sink

Start from how flow reaches a venue on Solana in 2026: aggregators run the auction. Essentially all of Omnipair’s swap volume arrives by router (Jupiter first), none of it organic. A router sends a fill to a static-fee, full-range pool only when that pool’s quote beats — at that instant — every oracle-anchored prop AMM quote and every volatility-priced DLMM quote on the network. A constant-product pool’s price only moves when someone trades it, so the moments it wins the auction are disproportionately the moments its quote is stale against a move that already happened somewhere else. The pool is a resting limit order that never cancels: it systematically sells what’s rising and buys what’s falling, for 25bps, to counterparties who know exactly why they’re there.

This is the loss-versus-rebalancing result (LP P&L = fees − a volatility-scaled arbitrage bleed) with the modern twist that makes it terminal: the pool no longer even receives the benign flow whose fees are supposed to fund the bleed. Retail-shaped flow gets filled upstream by whoever tops the auction; the passive pool gets the residue. Omnipair’s average fill is nine dollars. “Impermanent loss on the assets they’re trading well” is precisely the failure mode: the better an asset trends, the more the LP’s inventory is positioned the wrong way, and the more of its volume arrives specifically to take that inventory.

And the residue only gets worse with time. Venue-attributed swap data from last quarter shows prop-AMM share of an asset tracks one thing: whether it can be hedged somewhere else. Assets priced only on-chain round to zero; the moment one gains an external reference — a perp listing, a CEX market — its flow converts to prop AMMs within weeks, and the conversion keeps getting faster. So the pool’s best case is holding an asset through the phase where nobody wants size on it, then donating the flow to venues that take no public LPs the moment it matures. There is no phase where this pool is the venue of choice.

The yield can’t pay for it

The design’s answer to the above is the second yield source: lending interest on top of swap fees. The rate mechanics quietly cap that answer. Rates only ratchet when utilization crosses the model’s upper band; observed utilization sits comfortably below it, so every major pool’s borrow rate is pinned at the 1% floor — which at ~40% utilization pays the supplying LP about 0.4% a year on the lent side. Protocol-wide, LPs earned $345 in swap fees over the last seven days. Call the all-in gross yield mid-single-digits on the best pool — against full-range IL on tokens that reprice 2–5× as a matter of course, where a single 2× move costs ~6% and a 4× costs ~20%. One repricing eats years of the entire yield stack. The structural version of this point: borrow demand in a pair pool is leverage demand on that same microcap — the “second yield” is compensation for concentrating a second exposure to the exact risk the LP is already long.

Even the best case fails on its own terms: on META, its flagship pool’s asset, Omnipair fills under 6% of the day’s volume — the token’s own DAO-adjacent AMM and @MeteoraAG take the rest. And the LPs have noticed: two add-liquidity calls in the last week, protocol-wide.

Anyone sophisticated enough to LP volatile long-tail pairs profitably would do it where dynamic fees pay them for the toxicity — or run oracle-anchored quotes themselves. The set of people for whom Omnipair LPing dominates is the set who shouldn’t be market making at all.

The lending leg: unbundled by Offerbook

@JupiterExchange‘s Offerbook (public beta since May 27, 2026) is a permissionless P2P order book for USDC loans: fixed rate, fixed term (1–30 days), against essentially any on-chain collateral — verified tokens, xStocks, NFTs — with no price-based liquidations and no oracles. Terms hold for the life of the loan; default at maturity means the lender claims the collateral.

For the lender, compare the two ways to earn yield on dollars against long-tail collateral. On Omnipair you supply into a utilization curve that structurally can’t pay you (rates pinned at the floor unless the pool runs hot, at which point your exit is gated), you’re forced to hold the volatile side too, and you underwrite the pool’s write-offs. On Offerbook you name your rate, your LTV, and your term, per loan; your worst case is owning the collateral at your chosen discount. One of these prices tail risk; the other socializes it. The comparison isn’t close, and it’s the same collateral universe — Offerbook exists precisely for assets pooled-oracle lending can’t touch.

For the borrower — and observed demand on Omnipair is exactly this: short-duration dollars against volatile microcaps — the contract shapes differ in kind. An Omnipair borrow is a liquidation-market loan marked by an EMA whose downward leg snaps instantly; in a volatile market a keeper writes you off with a penalty. An Offerbook borrow cannot be liquidated by any price path during its term — Jupiter’s own docs list “borrow against a volatile asset and walk away if it craters” as a use case. A fixed-term loan is simply the better-shaped instrument for volatile collateral; “loans that can be liquidated in a very volatile market” is the product nobody needs to accept anymore.

Honest caveats: Offerbook is USDC-only, term-capped, and swaps price risk for maturity discipline; pooled lending still wins for instant perpetual credit on majors. But majors were never Omnipair’s market — Kamino, marginfi, and Jupiter Lend own that. The kernel of Omnipair’s lending TAM — permissionless credit on the long tail — is what Offerbook was built for, without conscripting an LP to hold a pool.

The Corner Cases

I went looking for the configurations where the GAMM should shine. Each one dissolves:

• Correlated pairs (the IL-free case): no divergence means nothing to leverage and no volatility to charge for — no borrow demand, no fees. Omnipair’s own LST pool has never had a borrow. The IL-free configuration is also the yield-free configuration, and LST liquidity has better homes anyway.

• A project bootstrapping its own token’s money market (the strongest story, and the actual observed user base): the flagship example fills a single-digit share of its own token’s volume on Omnipair while the token’s liquidity lives elsewhere. What bootstrapping buys is a thin USDC side at floor rates and tens of dollars a day in fees. A treasury that wants borrow-against-own-token posts it on Offerbook; one that wants liquidity seeds Meteora — which is where these projects’ liquidity already is.

• “Yield on assets nothing else will touch”: Offerbook’s collateral universe covers the tail from the lender side with explicit per-loan pricing — no pairing, no gating, no socialization.

The engineering earnestly mitigates the borrower-side abuse vectors (pessimistic EMAs, impact-aware collateral factors, isolation), and there’s been no exploit. But every safety mechanism is LP-funded: over-seize on noise or under-recover in a gap, the difference lands in the same place. Containment isn’t a yield.

Could this have worked earlier?

Probably — as recently as 2023. When passive xy=k pools were the only on-chain liquidity, fees on trending assets were fat and the arbitrage bleed was shared among slow arbitrageurs; when no permissionless fixed-term credit existed, utilization curves against thin collateral were the only game, and the Fuse era proved real demand for it. A GAMM in that world plausibly out-earned its parts.

Every pillar of that world is gone on Solana. Aggregators intermediate the flow and run the auction. A stable prop-AMM layer converts every asset the moment it becomes hedgeable and takes no public money. Dynamic-fee concentrated liquidity owns the discovery end and charges volatility for it. Fixed-term P2P credit unbundles the lending leg with the dominant router’s distribution behind it. Omnipair is a full-stack answer to a question each part of the market now answers separately, better. The window for “your idle AMM liquidity is also a loan book” closed before the protocol shipped.

Conclusion

Omnipair asks one LP position to lose two separate wars: against dynamic-fee concentrated liquidity and the prop-AMM layer for fills, and against fixed-term credit markets for lending. Its own documentation concedes each load-bearing risk — IL only “partially offset,” withdrawals utilization-gated, bad debt “socialized to LPs” — and the logic of the router meta guarantees the offsets can’t materialize: the pool’s fee income is capped by an auction it only wins when it’s wrong, and its lending income is capped by a rate curve its own thin demand never bends. The GAMM’s liquidity provider is the counterparty of last resort to everyone smarter, faster, or better-structured. And none of it is fixed by growth, incentives, or time — the failure is the position itself, so the only thing that would change my mind is a different design: volatility-priced fees, oracle-informed quoting for assets that have one, a credit leg that prices collateral instead of utilization. At which point it isn’t this protocol anymore. That’s why I can’t get bullish Omnipair.

Introduction

Everyone knows prop AMMs fill most of SOL/USDC. This paper is about the better question: what decides which assets they quote at all. I measured every venue-attributed swap leg on Solana over the last quarter — April through mid-July, ex-pump.fun, fourteen prop AMMs in the data, twelve still active into the final month — and the answer turns out to be one variable: whether the asset can be hedged somewhere else. That single property predicts how much of each asset’s volume prop AMMs fill, how fast new assets migrate to them, and which assets they’ll never touch. Three patterns below, then what they mean for where trading on this chain is going.

Pattern one: the easier an asset is to hedge, the more of its trading happens on prop AMMs

Line up every asset class by how much of its trading prop AMMs fill, and you get a strikingly clean gradient:

• Stablecoins and other par-redeemable assets: ~50% (only an order book competes)

• SOL/dollars: ~61%

• Wrapped ETH against stables: 96%; wrapped BTC: ~36–50%

• CEX-listed large caps: 38–68% (two-thirds of PUMP volume fills on prop AMMs)

• Tokenized equities: 32–44% of each name, ten weeks after the class was born

• Perp-listed memecoins: 17–44%

• Anything priced only on-chain: zero. Not “small” — zero.

That’s not a list of market-share outcomes; it’s a single variable expressed seven ways. Prop AMM share of an asset is a direct read on how cleanly the asset can be hedged somewhere else, discounted by how much of its flow arrives through routers rather than directly at pools. Nothing else — not age, not volume, not volatility — sorts the data anywhere near as well. The biggest memecoin volume periods of the year attracted zero prop AMM entry, while $8M/month FOGO trades 92% on prop AMMs because it has an external price.

The discount term matters as much as the rule. Wrapped ETH is 96% prop-AMM-filled when quoted in dollars but 42% overall, because Orca keeps the SOL-denominated flow. Prop AMMs own what routers route; pools keep what users bring directly. So the ceiling on prop AMM share of any asset = its hedgeability × its router-routed fraction. Both terms are rising: listings keep adding reference prices, and aggregators have been taking a growing share of Solana order flow.

Pattern two: an asset moves to prop AMMs as soon as it gets an outside price — and that move keeps getting faster

The time series says the frontier doesn’t advance by competition — it advances by qualification. SOL/stable’s prop AMM majority predates the window and didn’t move (66% → 61%). No established category flipped through head-to-head fighting this quarter. What changed is that new assets qualified, and qualified assets converted almost instantly:

• Tokenized equities: category born ~May, $2.2B/month by July, prop AMMs at 32–44% of the top names essentially from listing. Zero-to-co-lead in weeks.

• New stables: USD1, USDe, USDG onboarded on prop AMMs first — hundreds of millions a month on AlphaQ and Aquifer before any public pool got deep.

• New pair formats: the quarter’s one genuine innovation, Scorch quoting alts against SOL (the AMMs’ native pair format), took prop AMM share of BTC/SOL from 11% to 36% in weeks.

Meanwhile the fleet beneath this stayed churn-heavy and share-stable: three venues died or exited (ZeroFi, Obric, effectively Aatrox), one pivoted businesses entirely (Goonfi), two grew several-fold (Quantum 9×) — and fleet-level share of every settled category barely moved. Individual venues are interchangeable; the layer is what persists. Every per-token share “loss” this quarter traces back to a venue exiting, not a public AMM winning flow back — and every venue that grew had a deepening router relationship, while every venue that died had lost one.

Put those together and the machine is visible: a stable prop AMM layer, fed by routers, that converts each newly hedgeable asset on arrival — and, in this window at least, gave none back. The conversion is also getting faster: SOL’s migration predates the data, CEX-listed alts took months, tokenized equities took weeks — because the venues and the routing rails now pre-exist any new listing.

Pattern three: new tokens start on AMMs; once they mature, their trading ends up on prop AMMs

Follow an asset through the data and the two venue types stop being competitors and start being phases. A token is born on a launchpad or an AMM pool — the only place an asset with no external price can trade, because pools are price-discovery machines that don’t need a reference. If it dies there (most do), it never touches a prop AMM. If it matures — gets a perp listing, a CEX market, an underlying — its flow migrates to prop AMMs within weeks, and the pool keeps only the impulse trades and the SOL-denominated remainder. The tokenized-equity class is this whole lifecycle compressed: born with a reference price, it skipped the AMM phase almost entirely, and public pools never became its primary market.

That’s the future the data keeps pointing at. Public AMMs are becoming Solana’s primary market — the discovery venue where new assets find a price — while prop AMMs become its secondary market, where everything with a known price actually trades. The “DEX volume” people quote is the sum of a maturing secondary market and a casino-sized discovery layer, which is why the number swings 6× in a quarter while prop AMM flow holds within ±20%.

What this implies about trading on Solana in a year

New asset classes will skip public liquidity entirely. Equities did. RWAs, new stables, and anything born with a reference price will onboard on prop AMMs first, and deep public pools for them may simply never form. The AMM phase of an asset’s life is becoming optional — reserved for assets that need price discovery.

The stack converges on the equities-market shape. Streamed prop AMM quotes filling router flow, one order book holding the stables middle (Manifest), AMMs as the listing/discovery venue, aggregators as the brokers deciding whose quotes fill. The remaining question is the one equities answered with PFOF: whether the dominant broker stays neutral. Eight of twelve prop AMMs live on Jupiter’s auction; DFlow and OKX already run their own venues. If Jupiter ever does the same, the independent prop AMM fleet becomes a captive one, and the current competitive equilibrium was a phase.

Execution quality becomes the product, and opacity the price. Prop AMMs won the majors on quote freshness, and they publish nothing — no IDL, no events, no inventory. Solana is converging on markets that settle transparently but form prices in closed systems. Analytics built on pool state will describe a shrinking share of real trading.

The system’s stress behavior is now untested — and different in kind. Public AMM liquidity is locked and passive; it cannot leave in a crash. Prop AMM liquidity is a revocable business decision — this quarter, three venues’ worth of it vanished in calm markets. A majors market where 60%+ of volume fills on prop AMMs has never been through a real Solana stress event. When it is, the pools that still exist beneath the quotes are the fallback — and they are thinnest exactly in the assets prop AMMs took over.

Sources

All figures come from our own in-house Solana database — a full-chain index we build and maintain that records every DEX swap leg with venue attribution, rather than third-party dashboards. The window is April 1 – July 14, 2026, and excludes the pump.fun complex (bonding curves and PumpSwap), which the index does not cover. All three months are valued identically — each swap off its stablecoin leg at $1, otherwise its SOL leg at that day’s price — because native USD fields don’t exist for the earlier data. Note that per-leg totals include Meteora DAMM-v2 pool churn that curated series filter out; excluding Meteora, our figures match Blockworks’ DEX-spot series within ~6% on sampled days. Prop AMM shares are floors: a handful of micro venues are absent from the index.

Executive Summary

Q2 2026 turned on a single event: the late-May discovery of a critical, four-year-old soundness bug in Orchard — Zcash’s flagship shielded pool — that could have produced undetectable counterfeit ZEC. The disclosure cut ZEC nearly in half and, over the following two weeks, drained about 14% of shielded ZEC from the pools. It also set the network’s agenda for the rest of the year: an emergency hard fork, a new pool to replace Orchard outright, and an accelerated push toward NU7.

ZEC had rallied to multi-year highs near $650–670 in May, aided by Multicoin Capital’s disclosure of a large position. The bug news reversed it — a ~46% drop to the mid-$300s in early June, then a recovery to the mid-$400s by quarter-end. Shielded balances fell ~14.4% (about 745K ZEC), almost all of it leaving Orchard. The turnstile mechanism made supply inflation impossible, and the emergency NU6.2 hard fork closed the flaw within days with no evidence of exploitation. Orchard ended the quarter at ~85% of shielded supply.

Network fundamentals moved the other way. Hashrate rose ~38% to new all-time highs, climbing through the crash, while the number of mining pools winning blocks per day held in the mid-teens — roughly double a year earlier. Google search interest for “Zcash” more than doubled to its highest readings in five years outside the November 2025 peak. Shielded-pool throughput, net of price, ended the quarter flat: the early-June deshielding surge round-tripped, and native volumes stayed below their late-2025 pace.

The fix replaces Orchard rather than patching it. Ironwood — a new shielded pool built with formal verification (via the Lean theorem prover) and quantum-recoverability groundwork — requires every holder to migrate through the turnstile, which lets any node independently re-verify the full shielded supply once migration completes. Proposed jointly by all four core organizations and targeting mainnet in late July, Ironwood lands alongside the first NU7 testnet, which put Tachyon’s ~300% scaling gains on-chain. ZODL, the team that spun out of Electric Coin Company in January, wrote the fix — its second emergency response in two quarters.

Measured against Bitcoin, the quarter looks stronger still. BTC fell ~14% in Q2; ZEC gained roughly 75% against it, ending well above where it started even after the June drawdown. That relative strength, the speed of the core teams’ response — a coordinated fix and emergency fork within days — and the ambition of the Ironwood rebuild turned a potential existential counterfeiting scare into a demonstration of the network’s resilience and the energy of its participants. For a quarter that opened under a serious threat to the protocol’s core guarantee, the incident proved, on balance, bullish. Whether holders migrate cleanly into Ironwood is the central question for Q3.

Shielded Pools

Zcash distributes its total coin supply across five pools, each serving a different function: Transparent holds ZEC that behaves like Bitcoin, fully visible on-chain; Sprout and Sapling are legacy shielded pools that use progressively stronger zero-knowledge cryptography to hide transaction details; Orchard is the current-generation shielded pool, offering the strongest privacy and the most efficient proofs, with no trusted setup; and Lockbox is a governance-controlled reserve that accumulates a share of block rewards for community-directed spending. How supply is distributed across these pools is the single most important lens on Zcash’s network health, because it shows whether users actually adopt the privacy features that justify ZEC as a distinct asset — a network with 95% of supply in Transparent is just a smaller Bitcoin. In a normal quarter, migration from Transparent into shielded pools, and from legacy pools into Orchard, signals genuine demand for privacy and growing sophistication among holders. In a quarter marked by a security scare, the same numbers measure the depth of the shock: whether a flaw in one pool triggers a wholesale retreat from privacy or merely a defensive reshuffle across pools, and whether the outflows prove permanent or temporary. And because the size and turnover of each shielded pool set the strength of the anonymity set that protects everyone in it, tracking these figures over time shows not just whether Zcash is delivering on its thesis, but how resilient that thesis is under pressure.

Pool balances

In Q2 2026, ZEC held in shielded pools fell by roughly 14.4%, a net withdrawal of about 745K ZEC. The decline was driven not by issuance or ordinary flow dynamics but by panic: in early June, disclosure of a critical soundness bug in Orchard — an under-constrained element in its zero-knowledge circuit that could have permitted double-spending, or undetectable counterfeit ZEC, inside the pool — set off a wave of deshielding as holders pulled funds out. The scare proved contained: Zcash’s turnstile mechanism prevented any inflation of the total supply, and an emergency hard fork (NU6.2) closed the flaw within days with no evidence of exploitation. The outflow was concentrated in the implicated pool — Orchard contracted by nearly 753K ZEC, while Sapling grew about 8K and Sprout slipped roughly 15 ZEC. The quarter sharply reversed a multi-quarter expansion trend, though Orchard remained the dominant shielded pool at roughly 85% of shielded supply even after the withdrawals.

The exited ZEC did not leave the network — it resurfaced in the Transparent pool, which grew about 888K over the quarter (the deshielded coins plus new issuance), while the Lockbox reserve accrued its usual ~20K in block rewards. And because the withdrawals shrank Orchard without fragmenting activity across venues, its anonymity set thinned but stayed by far the deepest on the network — the privacy guarantee eroded at the margin rather than breaking.

Pool flows

In Q2 2026, the USD value of ZEC flowing in and out of shielded pools rose over the quarter, but almost entirely on ZEC’s price appreciation rather than more coins actually moving: on a native-token basis, throughput ended the quarter essentially flat, after a sharp early-June surge — roughly 2.4x April levels, driven by the deshielding rush out of Orchard around the bug disclosure — that had fully unwound by quarter-end. Beneath that flat surface the pools diverged: Sapling grew its native flow by about 30%, while Orchard’s throughput round-tripped — spiking with the June exodus, then falling back to roughly flat on the quarter. Both pools remain well below their late-2025 run-rate, with Orchard’s native flow roughly halved since Q4 2025 — a continued, if uneven, cooling in shielded-pool throughput even as ZEC’s price ran higher.

Network Attention

When looking at Google search trends — the clearest public proxy for external attention on the network — Q2 broke the pattern of the prior quarter. Rather than the consistent bleed that defined Q1, attention bottomed out near its post-pump lows through March and April, then re-accelerated sharply: search interest for “Zcash” more than doubled from roughly 22 in April to about 50 across May and June — the second- and third-highest monthly readings in the past five years, behind only the November 2025 all-time high. The rebound tracked ZEC’s ~60% price appreciation over the quarter, with the early-June disclosure of the Orchard soundness bug and the NU6.2 emergency hard fork adding a second leg to the news cycle. Searches for “Zcash” continued to run at roughly twice the volume of the “ZEC” ticker throughout. The quarter still sits well below November’s peak, but Q2 ended with materially more external attention on the network than it began — the first quarterly upturn in interest since the Q4 2025 pump.

Mining Metrics

In Q2 2026, the hashrate of Zcash picked back up, rising roughly 38% after Q1’s slight cooling and pushing network difficulty to new highs, with the steepest climb coming in June. The number of mining pools winning blocks per day held in the mid-teens through the quarter — around 14 per day in April, peaking near 15-16 in May, and easing to about 12-13 by late June. That is still roughly double the 7 pools per day seen in July 2025, so miner diversity has held near its highs rather than reverting. Most tellingly, this hashrate strength came right through a violent mid-quarter shock — ZEC fell about 46% from its ~$670 late-May peak to ~$362 in early June on the Orchard vulnerability disclosure before partially recovering — yet mining difficulty kept climbing and pool participation stayed elevated, so the continued growth in hashrate reflects deepening long-term commitment to the network’s infrastructure rather than any retreat by miners.

Network Development

The Orchard flaw — and the Ironwood pool replacing it

On May 29, researcher Taylor Hornby — auditing for Shielded Labs with a custom agent framework built on Anthropic’s Claude Opus 4.8 — found a four-year-old soundness bug in Orchard’s zero-knowledge circuit (halo2_gadgets). An under-constrained check let invalid inputs pass, meaning an attacker could have minted counterfeit ZEC inside the pool. The turnstile capped the blast radius — total supply could never inflate — but balance integrity inside Orchard could no longer be taken on faith. ZODL engineers (Daira-Emma Hopwood, Kris Nuttycombe, Jack Grigg) moved fast: a soft fork disabled Orchard on June 2 (block 3,363,426), and the NU6.2 hard fork re-enabled it with a corrected circuit on June 3 (block 3,364,600). No exploit was found — but because privacy makes counterfeiting impossible to disprove after the fact, a patch alone couldn’t fully restore trust.

Hence the quarter’s biggest development: Ironwood, a brand-new shielded pool that replaces Orchard outright. Rather than keep patching, developers are sealing the legacy Orchard pool — no new deposits or internal transfers — and having every holder migrate funds through the turnstile into a fresh, corrected pool. The mechanic is the whole point: because the turnstile enforces that no more value can leave Orchard than verifiably entered it, once migration completes any full node can independently confirm that no counterfeit ZEC crossed into Ironwood — restoring trustless supply verification at the protocol level. Wallets will prompt the move in roughly a single tap.

Ironwood is also a bet on stronger assurance. It’s being built with formal verification — contributors from zkSecurity and ZODL are using the Lean theorem prover to mathematically prove the soundness of the Action circuit, alongside conventional audits and AI-assisted review — a direct answer to a bug that showed code audits alone weren’t enough. It also lays the groundwork for quantum recoverability, a smoother future path to post-quantum security. Proposed jointly by ZODL, Tachyon, the Valar Group, the Zcash Foundation, and Shielded Labs, Ironwood targets mainnet around July 21, 2026, with legacy zcashd support ending near block 3,417,100.

ZODL

The ex-ECC team that went independent in January — taking the wallet (now Zodl) and closing a $25M+ seed in March — spent Q2 executing. It shipped through Zodl v3.6.0 (June 17), adding worldwide currency conversion and local-currency display for its growing Brazil and Mexico bases, and previewed “quantum-recoverable” wallets in May. But its defining contribution was, again, incident response: ZODL engineers authored the NU6.2 Orchard fix and now co-lead Ironwood and Tachyon’s testnet integration. Two clean emergency patches in two quarters (Sprout, then Orchard) have cemented the young team’s operational credibility.

Zcash Foundation

ZF kept the node stack moving. Zebra reached 5.2.0 (June) with stronger consensus-split protection, a 10× larger reorg window, and signed, automated releases, while the Z3 stack advanced toward replacing zcashd — Zaino 0.4.0 hit full Lightwalletd parity and Zallet Alpha 4 added zcashd wallet migration, the on-ramp for the Ironwood cutover. ZF also published its Q1 report (~$817K spent), pushed FROST toward v3, and announced Zcon7 — Oct 27–29 in Cancún, timed to Zcash’s 10th birthday.

Shielded Labs

Shielded Labs had its most consequential quarter yet — not from a product, but because the audit it funded caught the Orchard flaw, and it co-authored the disclosure and coordinated the response (a repeat of its Q1 Sprout role). On its own roadmap it completed Crosslink Milestone 5 (April 15), pivoted to long-running “seasonal” FeatureNets, ran two workshops hardening fork-recovery and rollback, and kept prototyping the PoS slashing tooling. It also continued stewarding the Network Sustainability Mechanism (ZIPs 233/234/235) as an NU7 candidate.

Tachyon

For Sean Bowe’s scalability project, Q2 was the shipping inflection. The first NU7 testnet went live May 22, putting Tachyon on-chain and demonstrating a ~300% throughput gain — doubling shielded TPS and cutting block times from 75s to 25s, en route to a target of thousands of TPS. The quarter also sharpened its signature idea, Oblivious Synchronization (wallets fetch only the proofs relevant to them from untrusted servers, revealing nothing), and earned CoinDesk Research coverage plus >90% ZCAP/coinholder support. A year on from Bowe’s proposal, Tachyon moved from whitepaper to imminent upgrade.

What to Watch in Q3

• Q2 closed with the network mid-pivot: the Orchard episode contained, but not fully resolved until holders complete the migration to Ironwood. Five threads will define the quarter ahead.

• Ironwood migration. Whether holders move cleanly through the turnstile into the new pool — and the moment migration completes and any node can re-verify the full shielded supply, formally closing the Orchard episode.

• NU7 mainnet. Activation (targeted for late July) and whether Tachyon’s ~300% throughput gains hold up in production, not just on testnet.

• Shielded rebuild. Whether the ~745K ZEC that deshielded returns to privacy inside Ironwood or stays transparent — the clearest read on whether June was a permanent scar or a passing scare.

• Governance. Final NU7 scope, including the Network Sustainability Mechanism (ZIPs 233/234/235) and Zcash Shielded Assets.

• zcashd sunset. The legacy node’s end-of-support near block 3,417,100 and the Z3 stack (Zebra + Zaino + Zallet) taking over production infrastructure.

Executive Summary

Solana has effectively won the market for tokenized equities, capturing roughly 97% of all tokenized‑equity spot trading volume in May 2026 (~$869M on Solana vs. ~$24M across every other chain combined). But the four issuers that define the landscape — Backpack Securities (SPCX), Ondo Global Markets, xStocks (Backed Finance), and PreStocks — are not the same product wearing different logos. They look identical on a price chart and are radically different legal instruments underneath, sitting on a spectrum from redeemable into the real security to purely synthetic private‑market exposure. The distinction is invisible until something breaks — and in May 2026, with PreStocks, something did.

What a holder actually owns, ranked best to weakest:

• Backpack / SPCX — closest to owning the real thing. The token is redeemable back into the underlying share via ACATS/DTCC rails for eligible holders, landing on a UCC Article 8 security entitlement with genuine ownership rights. The catch: redemption is gated to onboarded holders; the raw on‑chain token is still an SPV claim.

• Ondo Global Markets — the best‑protected non‑redeemable model. A structured note against a bankruptcy‑remote SPV, 1:1 + buffer collateral, a third‑party security agent with a first‑priority interest, and daily verification — but explicitly no shareholder rights.

• xStocks — the most DeFi‑native option (Raydium, Jupiter, Kamino), but a bearer‑debt tracker certificate with no shareholder rights, and — contrary to common marketing — collateral that “may not always consist of the underlying shares.” Holders carry Backed’s credit risk.

• PreStocks — the weakest posture: synthetic exposure to pre‑IPO SPVs with no rights and disputed backing. In May 2026, Anthropic and OpenAI called the underlying share transfers void/invalid and the tokens fell 34–40%; PreStocks showed an implied ~$1.3T Anthropic valuation against only ~$23M in actual assets, and the attestation reports promised at launch were never published.

The decisive variable for on‑chain pricing is arbitrage: a tokenized stock carries two prices — the 24/7 on‑chain price and the underlying share value set by a market that closes nights and weekends — and when a live underlying market exists to arbitrage against (xStocks, Ondo, and Backpack during US hours), pegs hold tight and typical swap slippage tolerance on liquid names is just ~0.1–0.5%, whereas off‑hours, on weekends, or for PreStocks where there is no public underlying at all, premiums and discounts open up and prices can dislocate violently; liquidity is also heavily concentrated in a few marquee tickers (TSLAx, NVDAx, CRCLx, SPCX), so the 100‑plus long‑tail names quote wide and slip hard on any real size. The bottom line is that Solana has won the venue and the open question is which structure wins the holder — and regulation is already tilting the answer, with the January 2026 SEC staff statement’s practical effect being to pressure purely synthetic models and favor issuer‑sponsored and redeemable‑custodial ones, so for anyone choosing among these tokens it is the legal wrapper, not the ticker, that determines what you’re actually holding.

Background

A tokenized stock is a crypto asset whose ownership is recorded on a blockchain and whose value references an equity. The SEC staff statement of January 28, 2026 (Divisions of Corporation Finance, Trading and Markets, and Investment Management) established the governing taxonomy. It distinguishes:

• Issuer‑sponsored tokens — can represent true equity ownership.

• Custodial / entitlement tokens — a third party holds the real share and issues a token evidencing an indirect interest.

• Synthetic / linked tokens — economic exposure only, no claim on a share.

The statement’s key warning: third‑party tokenized securities “may provide different rights” than the underlying, and holders may face risks specific to the tokenizer (e.g., its bankruptcy) that a direct shareholder would not. This is the lens for everything below.

The Four Offerings

Backpack Securities — SPCX

Issuer & mechanics. SPCX is issued by Backpack Securities, a regulated US broker‑dealer, and launched on Solana June 11–12, 2026 to coincide with SpaceX’s Nasdaq IPO — reportedly the first time a newly listed equity had a simultaneous on‑chain market. Each token is backed 1:1 by a real SpaceX share purchased and held in regulated custody. Liquidity is routed by Sunrise DeFi.

Legal structure & holder rights. Backpack’s structure is the strongest for holders because the token is redeemable back into the underlying equity. Eligible (onboarded, KYC’d) holders can redeem tokens for the real share and transfer it to any traditional brokerage via ACATS/DTCC settlement rails — the same infrastructure that moves stock between Schwab and Fidelity. On the brokerage side this is a UCC Article 8 security entitlement with real ownership rights (dividends, corporate actions, transfers). Caveat: the raw on‑chain token is still, in Backpack’s own description, a “tokenized claim / claim on an SPV.” Redemption into the real share is gated to eligible holders — a random secondary‑market buyer who never onboards is closer to the note holders below.

On‑chain market data — spread & value. SPCX surpassed $100M in 24‑hour volume by June 15, 2026, accounting for roughly 40% of all Solana tokenized‑equity trading in its first days. The underlying SpaceX equity priced at $135 (IPO), closed day one at $160.95 (+19%), and hit an intraday high of $225.64 on June 16. Because SpaceX was newly listed and volatile, the tokenized form carried a meaningful new‑listing / crypto‑access premium: tokenized variants (e.g., BitMart’s bSPCX ~145 USDT, third‑party SPCXON) deviated from the Nasdaq price, especially during after‑hours and crypto‑only sessions when no live equity market exists to arbitrage against.

Risks. Concentration (a single name so far), redemption gated to eligible holders, and the standard custody/operational risk of the broker‑dealer.

Ondo Global Markets

Issuer & mechanics. An Ondo tokenized stock is a structured note — a debt instrument issued by Ondo Global Markets (BVI) Limited, a bankruptcy‑remote SPV. Launched January 2026, Ondo now lists 264 tokenized equities and ETFs and reports >$1B in total value locked — the most extensive equity/ETF roster of the four. Tokenholder rights are governed by Swiss law.

Legal structure & holder rights. This is the best‑protected non‑redeemable model. Tokens are backed 1:1 plus a buffer by underlying securities held at regulated custodial broker‑dealers; Ankura Trust Company acts as Verification and Security Agent, holding a first‑priority perfected security interest in the collateral, with daily verification, monthly reconciliations, segregated accounts, an independent director, and annual audits. But the rights gap is explicit and quoted verbatim from Ondo’s legal docs: “you will not see your name on the share register” and “you do not have shareholder voting rights, shareholder information rights or other shareholder rights.” A Broadridge integration lets holders express voting preferences Ondo may apply — a governance‑experience feature, not ownership.

On‑chain market data — spread & value. Ondo led the tokenized‑stock market by value — roughly ~58–60% of the tokenized‑stock market on RWA.xyz in early–mid 2026 — with ~264 listed assets and >$1B TVL. Because Ondo’s model is collateral‑backed and institutionally arbitraged, on‑chain prices track underlying NAV closely during US market hours; deviations appear off‑hours like all the others.

Risks. Holder owns a note claim against an SPV, not a share — no shareholder rights; exposure to issuer/custodian operational failure (mitigated, not eliminated, by the security‑agent structure).

xStocks (Backed Finance)

Issuer & mechanics. xStocks are SPL tokens on Solana issued by Backed Assets (JE) Limited (Jersey), launched June 30, 2025. The lineup spans 130+ equities and ETFs — AAPLx, TSLAx, NVDAx, METAx, GOOGLx, COINx, CRCLx, MSTRx, plus SPYx and QQQx. Each is structured as a bearer debt instrument classified as a tracker certificate. xStocks is the most DeFi‑native option: Raydium is the primary AMM, Jupiter aggregates quotes, and Kamino accepts xStock collateral for borrowing.

Legal structure & holder rights. Economic exposure only — no shareholder voting, no direct dividend rights (dividends passed via a rebasing mechanism), no legal claim to the underlying shares or to residual assets in a company liquidation. Important correction to common marketing: xStocks is often described as “fully 1:1 collateralized in the underlying shares,” but xStocks’ product disclosures (surfaced in Kraken’s and Bybit’s risk pages) state collateral “may not always consist of the underlying shares” and “other eligible assets (including cash collateral) may be used as substitute collateral.” Holders bear Backed’s credit and solvency risk irrespective of the underlying’s performance.

On‑chain market data — spread & value. xStocks reached ~$293.5M AUM on Solana by mid‑May 2026, with >$3B cumulative on‑chain volume and >$517M DEX volume by early 2026. Liquidity concentrates in a few names (TSLAx, NVDAx, CRCLx); on those, recommended swap slippage tolerance is ~0.1–0.5% on Raydium/Jupiter. Spread mechanism: during US market hours, arbitrage between the on‑chain pool and the real equity keeps the peg tight; long‑tail tickers have thin books and wide spreads / high slippage for size; on weekends and after‑hours, with TradFi closed, prices float on pure crypto supply‑demand and premiums/discounts open up.

Risks. Issuer credit risk, substitute‑collateral risk, thin long‑tail liquidity, and no shareholder standing.

PreStocks

Issuer & mechanics. PreStocks tracks pre‑IPO private companies (e.g., OpenAI, Anthropic, SpaceX pre‑listing) via tokens backed by SPV exposure to private equity. Public volume figures are thin and unreconciled: the highest datable third-party daily volume is ~$29M (April 2026), while platform-cited figures (a $54.43M daily ATH, >$750M cumulative) remain unconfirmed by any third party. Operated under Regulation S — not available to US persons (and several other jurisdictions).

Legal structure & holder rights. The weakest holder posture. Tokens grant economic exposure only — no ownership, voting, dividend, or information rights. The claim depends entirely on the enforceability of upstream SPV interests in private companies with strict transfer restrictions. PreStocks itself disclaims being a broker‑dealer, adviser, exchange, transfer agent, custodian, or VASP.

On‑chain market data — spread & value. This is the most dislocated of the four because there is no public underlying market to arbitrage against — the “price” is a platform/implied mark, not an arbitraged NAV. The structural fragility became concrete in May 2026, when Anthropic and OpenAI publicly stated that the pre‑IPO SPV share transfers underlying these tokens were void/unauthorized (Anthropic: transfers to an SPV are “void under our transfer restrictions”), and the affected tokens fell 34–40% (Anthropic −34% over 7 days, OpenAI −39%; roughly −40% intraday). Compounding this, the platform showed an implied Anthropic valuation above $1.3T against only ~$23M in total assets, and no third‑party attestation reports promised at launch have been published. Spreads/discounts here reflect not just liquidity but genuine doubt about whether the backing is enforceable.

Risks. Disputed/possibly unenforceable backing, no attestations, no shareholder rights, illiquid private underlyings, US‑person exclusion.

On‑Chain Spreads & Share Value

Why spreads exist at all. A tokenized stock has two prices: the on‑chain trading price (set by DEX/CEX supply and demand 24/7) and the underlying share value (set by the equity market, which is closed nights and weekends). The gap between them is the story.

During US market hours — arbitrageurs (or authorized participants who can mint/redeem against the real share) keep the on‑chain price ≈ underlying. Pegs are tight; on the liquid names, typical swap slippage tolerance is ~0.1–0.5%.

Off‑hours and weekends — the equity market is closed, no fresh NAV anchors the token, and price floats on pure crypto demand. Premiums and discounts open up — most dramatically around events (a hot new listing like SPCX, or a pre‑IPO name reacting to news).

Liquidity is concentrated. Across all four issuers, volume clusters in a handful of marquee tickers (TSLAx, NVDAx, CRCLx, SPCX). The long tail of 100+ listed assets often has thin order books, meaning wide quoted spreads and steep slippage for any meaningful size.

Conclusion

The January 2026 SEC staff statement tightened scrutiny on synthetic equity while opening a supervised path for true entitlement models — a December 2025 DTC no‑action letter preserves UCC Article 8’s indirect‑holding framework. The trajectory favors issuer‑sponsored and redeemable‑custodial models, such as Backpack and Superstate’s native equities, and pressures purely synthetic ones like PreStocks. The PreStocks SPV‑transfer dispute is the clearest live example of the tokenizer‑specific risk the SEC warned about: the token traded fine until the referenced companies said the backing wasn’t valid.

That regulatory line maps directly onto how the four products actually differ. On a chart they all track a stock price, but legally and structurally they are four different instruments — Backpack is closest to owning and redeeming the real share, Ondo is the best‑protected note, xStocks is the most DeFi‑native tracker certificate, and PreStocks is the most speculative synthetic SPV exposure. On‑chain, the decisive variable for spreads is whether a live underlying market exists to arbitrage against, which is exactly why the liquid xStocks names hold a tight peg during market hours while PreStocks can dislocate violently. The venue question is settled; the structure question is not — and that, not the ticker, is what every holder is actually choosing between.

Introduction

@ManifestTrade is an onchain orderbook that supports limit and market orders. since the start of the year it has grown from under 1% of Solana spot DEX volume to a consistent 5-8% share, a notable trajectory in a period when proprietary AMMs are dominating overall flow. Total Solana DEX volume has actually contracted over the same window, so Manifest’s 5-8x growth in market share has translated to only a 2-3x increase in absolute volume. The fact that it is gaining share against prop AMMs rather than losing it is the interesting data point.

What separates Manifest from prior Solana CLOBs like Phoenix and OpenbookV2 is a redesigned cost structure for makers. Market creation rent is roughly 500x cheaper (~0.004 SOL versus venues that pre-allocated maximum order space at launch), compute per order is ~45% lower than Phoenix, and Global Orders let the same tokens back resting bids across unlimited markets simultaneously, with funds moved just-in-time at fill. The historical problem with onchain orderbooks on Solana was the economics of being a passive maker. Manifest fixes that, and the impact is sharpest on correlated pairs: a single bucket of SOL can simultaneously quote bonkSOL/SOL, jitoSOL/SOL, mSOL/SOL, and every other LST market without locking capital per venue.

Protocol Use

Two segments account for nearly all of its flow:

• Stablecoin-to-stablecoin swaps

• LST-to-SOL swaps

There is marginal activity in tokenized assets, but Manifest is essentially absent from memecoins, project tokens, and bridged majors. Its volume share is concentrated in a specific segment of the Solana DEX landscape: correlated pairs. This is not an accident, it reflects a structural fit between the orderbook model and the microstructure of correlated asset trading. Understanding why Manifest wins here, and why it does not win elsewhere, requires looking at what each venue type is actually selling.

Prop AMMs like HumidiFi, GoonFi, Obric, and SolFi monetize a pricing edge. They pull Pyth oracle data, quote around fair value, and manage inventory in real time. Their advantage is the ability to re-price continuously, which matters enormously when the underlying asset is volatile and informed flow is racing to pick off stale quotes. Passive AMMs monetize price discovery itself. They accept that LPs will be adversely selected in exchange for fees on high-volume speculative trading, which works for the long tail of tokens where no reliable oracle exists.

Manifest wins where neither of those models has an edge. On correlated pairs such as JitoSOL/SOL, mSOL/SOL, and stablecoin pairs. The JitoSOL/SOL exchange rate is a deterministic function of the staking redemption rate. USDC/USDT trades in a tight band anchored by redemption. There is no proprietary pricing signal a prop AMM can extract from Pyth that a CLOB maker does not already have access to through the same public information. The prop AMM’s core monetization mechanism, an information advantage over passive LPs, collapses when the pricing function is public and slow-moving. In that regime, the orderbook’s ability to host multiple makers competing on price-time priority compresses spreads below what any single prop AMM is willing to quote, and Manifest’s low cost structure lets those makers actually show up.

Growth Outlook

The open question is whether Manifest extends from here. The flow it has captured is defensible because the underlying microstructure of correlated pairs is structurally stable. Stablecoins will remain stable and LSTs will remain anchored to their staking rates. Growth from this base requires either expanding the set of pairs where CLOB economics work, which is hard given the prop AMMs’ lock on liquid uncorrelated flow, or capturing a larger share of the correlated pair segment itself as overall Solana DeFi volume grows. The second path is the more likely near-term driver, and it is the reason Manifest’s current 5-8% share is worth watching rather than dismissing as a niche.

Executive Summary

ReFi Hub tokenizes revenue-share stakes in operational solar plants, with 18 months live, $491K deployed, 14% blended IRR, and zero defaults. The team is raising 14% of the company on Craft at a $4M–$8M valuation band. The forward bet is Volt — a composable wrapper token — and a pivot toward fund managers writing $2M–$20M tickets.

The most important variables for investors are whether the team can preserve deal yield quality while scaling, whether Volt achieves integration into major DeFi protocols, and whether the institutional capital pipeline converts. The structural tailwind is that solar revenue is a novel, off-chain yield source being bridged to Solana, largely uncorrelated with other on-chain yields, and the team has built underwriting credibility through 18 months of zero defaults.

The raise is best understood as growth capital bridging a working pilot to the institutional scale the asset class requires. The supply side — roughly $35–40M in assets in diligence — is largely in place; the demand side is early but no longer hypothetical — a signed LOI with TruFin plus an active fund manager pipeline — and the compliance and legal buildout this round funds is what converts it. At the top of the valuation band, a 2–6x return requires scaling from $491K deployed to roughly $100M, and a genuine home run requires $1B+. The bet, simply put, is on whether the team can scale solar sourcing by orders of magnitude while holding a yield premium wide enough for Volt to function as productive DeFi collateral. Everything else — fund manager conversion, protocol integration, return outcomes — flows from that spread.

The Product

ReFi Hub operates as what the team calls a “partner-manager-operator” model. They source solar plant operators — typically those holding tens to hundreds of millions in solar assets — diligence individual plants, and structure revenue-share agreements between the plant and onchain investors. Investors buy tokenized stakes in a specific plant’s revenue, see the plant’s historical performance and projected yield on the platform, and receive USDC distributions onchain as the plant sells electricity.

The structures are closer to equity than debt — investors hold a fixed percentage of revenue rather than a fixed coupon — though the team has been piloting a hybrid model. Their newest listing, FLP2, combines a fixed monthly payment (e.g., $10K/month) with a smaller variable revenue share (~20% rather than ~80%), which delivered roughly a 21% IRR over a shorter time horizon. Ownership stakes are capped below 60% of any given plant so operators retain majority control and the asset continues yielding for them.

Rooftop solar has been the primary focus during the pilot phase. Primary markets are Singapore and Brazil, with planned expansions into Japan, Malaysia, Korea, Indonesia, and New Zealand.

Track Record to Date

• Time live: ~18 months

• TVL: $491K

• Live Operating Assets: 7

• Blended IRR: ~14% across projects

• Defaults: zero

• Asset pipeline: ~$35–40M in various stages of diligence

The numbers are small in absolute terms. The pilot was deliberately structured around smaller deployments with trusted developer partners to build trust and an initial user base before scaling.

Major Initiatives

Two initiatives define ReFi Hub’s roadmap, and both are direct responses to the same structural constraint: retail KYCed USDC aggregation cannot scale to the deployment sizes that solar asset operators actually need.

Volt aggregates yield from solar assets across the platform into a single fungible token whose price appreciates over time, targeting a smoothed APY in the ~7–9% range. It does three things at once: weekly withdrawal windows and a USDC reserve provide a liquidity layer, the wrapper structure removes KYC gating at the depositor level, and the diversified yield profile gives DeFi protocols a single instrument they can underwrite as collateral. Individual revenue-share tokens tied to one plant cannot be listed in a money market. A diversified wrapper with predictable yield mechanics can.

The second is the pivot toward fund managers as the primary demand side of the platform. The thesis: plant operators with tens of millions in assets to list cannot wait on retail aggregation. The realistic ticket size required to move these deals is $2M to $20M per allocator, not $100K scratched together across hundreds of retail wallets. The features driving major interest are cross-border access to fragmented geographies and the operational overhead reduction that comes from smart contract automation replacing manual banking workflows. This pivot is no longer purely conceptual: the team has signed a letter of intent with TruFin, an institutional yield protocol, covering $5–10M in deployments, with five further fund manager conversations active at $2–20M each. The deeper detail and the caveats sit in Our Thoughts; the point here is that the demand side is moving from thesis toward signed paper.

The Raise

ReFi Hub is raising on Craft via sealed-bid auction. The mechanics:

• Tokenization: 14% of the company

• Valuation band: $4M to $8M

• Bid mechanism: sealed-bid auction. Participants submit private bids within the valuation range, and the clearing price is determined by Craft's auction mechanism

• Resulting raise size: between ~$400K and ~$1.12M, depending on the clearing price and the amount of equity ultimately tokenized — a smaller tokenized share lowers the floor below the ~$560K that a full 14% at a $4M cap would imply

Why Craft?

The team evaluated other on-chain raise platforms — MetaDAO was specifically considered — and chose Craft for its partial tokenization structure. Tokenizing 14% rather than the full company preserves flexibility for follow-on secondary rounds, a second Craft raise, or an organic path to breakeven without further dilution. The broader shareholder base also functions as a user acquisition channel during the raise itself.

Use of Proceeds

Three priorities:

1. Accelerating Volt launch — getting the single-token wrapper product to market faster and in front of a larger audience.

2. Fund manager conversion — Proceeds will fund the compliance buildout and legal infrastructure needed to check the specific boxes each fund manager requires.

3. Team expansion — modest headcount additions to support the above.

The raise is best understood as growth capital to bridge the gap between a working pilot ($491K deployed, 14% IRR, zero defaults over 18 months) and the institutional scale the asset class actually requires. The supply side — roughly $35–40M in assets at various stages of diligence — is largely in place. The demand side — the fund manager pipeline — has produced one signed LOI (TruFin, $5–10M) and several active conversations, but converting that interest into deployed capital requires the compliance and legal work this raise funds. That conversion work is what the raise is funding.

Our Thoughts

Why Volt Matters

The current product offers competitive yield, but the structure has two real ceilings: each deal is a fragmented standalone offering rather than a fungible position, and participation is gated behind KYC. Volt is the unlock for both. A single standardized yield token with permissionless distribution and DeFi composability is what grows the demand side by multiples rather than increments. The architectural pattern is becoming the default for long-tail RWA — Maple’s syrupUSDC is the closest structural analog on the credit side, and Onre is running the same playbook for reinsurance cashflows. ReFi Hub is applying it to solar.

The Revenue Math

The revenue math at scale is worth walking through. At $100M in cumulative assets deployed, the team would have generated roughly $3M in lifetime deployment fees at the 3% rate. If they can hold a 2–3% yield share on the underlying assets, that adds another $2–3M annually in recurring revenue. Call that a $3.5M annual revenue run rate at $100M AUM. At a 3x multiple, that’s roughly a $10.5M valuation. At a 10x multiple — defensible if the recurring yield share is treated as durable, DeFi-integrated cashflow rather than deal-flow revenue — that’s closer to $33M.

The Scaling Path

For investors entering at the top of the Craft valuation band ($8M cap), a 2–6x return requires the team to scale from $491K deployed today to roughly $100M deployed. That's a 200x increase in AUM, but the path runs through discrete institutional tickets rather than retail aggregation. The team has signed a letter of intent with TruFin, an institutional on-chain yield protocol that has introduced over $300M across digital asset and structured products since 2023. The LOI covers $5–10M in deployments into FIT and/or Volt structures, commencing Q3 2026 over a 24-month window, with three prospective allocators already identified from TruFin's network. Beyond TruFin, five additional fund manager conversations are active at various stages, each exploring $2–20M deployments, though no further LOIs are signed yet. Read against the $35–40M in late-stage supply-side diligence, the scaling path is a function of converting a handful of large tickets, not onboarding thousands of wallets. A genuine home run scenario — the kind of valuation that makes this category bet pay off — likely requires $1B+ deployed, which is closer to 2,000x from here.

The Core Question

The real question reduces to this: can this team source $100M to $1B in solar assets and package them into a yield product that delivers a return well above prevailing onchain lending market rates? The yield premium matters specifically because Volt’s collateral utility depends on it. If the token yields meaningfully more than the cost to borrow against it, looping works and demand compounds. If the spread compresses, Volt becomes just another wrapped yield product without a composability flywheel.

At launch, this trades sideways to down. Two setups would change the call:

Drawdown with thesis intact. If the token reprices materially below launch but pipeline progress, deal performance, and Volt timing remain on track, it becomes a buy on the dislocation rather than the thesis.

Momentum at flat FDV. If Volt adoption starts showing real numbers — AUM building, money market integration confirmed, named institutional deposits — and the token hasn’t yet repriced upward to reflect it, that’s the asymmetric entry.

The Project

Jurassic Finance is a Solana project proposing to acquire authenticated dinosaur fossils, place them inside Cayman SPVs, and sell fractional tokens against each one. The Labs entity, raising now under the ticker $RAWR, would operate the platform. The first fossil vehicle, a Triceratops head targeting roughly $1M, is scheduled for Q3 2026.

$RAWR holders own the operating company. They earn platform-level fees from every SPV that launches, receive 5% of each SPV’s token supply, and govern the treasury through futarchy. They do not own any fossil. They own a claim on the team’s ability to repeatedly originate and run fossil SPVs. SPV token holders ($TRCH1 being the first) own fractional economic rights to the underlying specimen via the Cayman entity. They benefit from the asset appreciating. They do not benefit from the platform succeeding beyond their specific fossil.

The risk profile of $RAWR depends almost entirely on execution. The risk profile of an SPV token depends on the specimen and the market for it.

What the Team Has Done

What exists: A website (jurassic.finance), a Twitter account with under 3,000 followers, an investor deck, a few thesis articles published by the team, a waitlist that the team reports at 3,300 signups, and a Cayman SPV structure that the team says has been scoped with legal counsel. The team has filed for Colosseum Frontier with a May 11 deadline. None of the contracts are deployed yet. No fossil has been acquired. No museum loan agreement has been signed. No SPV has been structured beyond the planning stage.

What the team has done elsewhere: Vukan runs a Solana creative studio called Blockformer that has produced marketing content for MetaDAO Ownership, Streamflow, Jupiter, and Solflare. Sora was previously COO at Blockformer and spent three years in marketing at LandX, a farmland tokenization project on Ethereum. The team’s combined background is content marketing and Solana-native community building. It is not asset acquisition, custody, legal structuring, or auction-house relationships.

What is claimed but unverified: Inbound from “100+ potential advisors, investors, and funds.” Active conversations with auction houses. A merchandise partnership the team has not named. An advisory bench that reportedly includes legal, Sotheby’s, tokenization, and paleontologist references.

This is not unusual for a pre-launch raise. Most projects raising on MetaDAO have a similar profile: a thesis, a deck, a small team, and a set of relationships. The relevant point is that none of the operational pieces that make the fossil tokenization model work have been built or tested yet.

Where the Qualitative Case Holds Up

The category is genuinely empty: Wine, watches, sneakers, cards, classic cars, fine art, and rare books all have at least one major tokenization platform. Fossils have zero. This is verifiable and it is structurally interesting. It either means the team has identified a real gap or it means the gap exists for reasons that make the business hard to build. Both are possible. The case for “real gap” is that fossils were not a serious institutional asset class until the last five years and the infrastructure simply has not caught up. The case for “hard to build” is that fossils are illiquid, opaque, hard to authenticate, and legally complicated in ways that wine and watches are not.

The auction market data is real: Ken Griffin paid $44.6M for the Apex Stegosaurus at Sotheby’s in July 2024, against a pre-sale estimate of $4M to $6M. Stan the T-Rex cleared $31.8M in 2020. These are documented sales. The bid for top-tier specimens has clearly stepped up. What is less clear is whether the demand at the very top of the market translates into demand for the $500K-to-$5M specimens that a tokenization platform would realistically be able to acquire.

The museum loan structure is sound in principle: Major museums do accept long-term loans of high-quality specimens and do typically assume conservation and insurance costs in exchange for display rights. This is how a lot of high-value paleontology specimens are housed today. Whether an accredited natural history museum will accept a loan of a specimen that is being publicly traded as a fractional security is an open question. The QR-code-to-fractional-ownership detail in the team’s pitch is the kind of thing that institutional curators tend to push back on. The structure works in concept. The execution depends on a museum being willing to participate.

The MetaDAO mechanism is genuinely protective: The launch valuation gets capped regardless of demand, the treasury operates under futarchy with a published monthly allowance, and the team’s tokens are locked for an 18-month cliff with price-milestone-gated unlocks after that. The structural protections for $RAWR holders are real, well-documented, and have been tested across the platform’s prior launches.

Where the Qualitative Case Gets Weaker

The five revenue streams look like diversification but on closer reading they all collapse into the same underlying requirement: the platform needs to tokenize a lot of fossils and build a sizable active buyer pool for SPV tokens. The origination fee and the 5% SPV token allocation are effectively the same line, both paid at issuance and both scaling with the cadence and size of SPV launches. The lending market requires a deep pool of SPV tokens on-chain to lend against before it generates substantive revenue, which is downstream of that same issuance cadence.

The FOSSIL buyback program has a different but related problem. The standing bid only gets triggered when on-chain demand for the SPV tokens is weak, which means the program activates exactly when the platform is struggling, not when it is succeeding. Each token the Reserve absorbs is then sitting on the Labs balance sheet until the team can find an off-chain buyer to take the inventory at or above the absorbed cost. If those buyers exist in size, the buyback program is a reasonable liquidity facility. If they do not, the Reserve compounds illiquid inventory while the platform’s headline demand is already weak.

The merchandise line reads as immaterial relative to the rest.

The “no competition” framing is partly an artifact of category immaturity, not an obvious moat: If Jurassic Finance is the first to demonstrate that fossil tokenization works at scale, the team has shown a competitor that the category is fundable. The actual moat is the relationship with auction houses and museums, neither of which the team has demonstrated publicly. A second mover with stronger institutional relationships and worse marketing could plausibly displace the first mover here.

The team comp set is content marketing, not asset operations: This is the single biggest structural concern. Buying, custodying, transporting, insuring, authenticating, and eventually selling a $1M to $5M physical asset is not the same skill set as running a Solana token launch. The team has not done it before and is not partnered with anyone who is publicly known to have done it before. The advisory roster may close this gap but the public-facing roster does not yet.

The Solana retail base for $1M-plus fossil SPVs is speculative: The Labs raise is filled by people willing to buy a $516K-FDV ownership token of a fossil platform. That is a fundamentally different buyer profile from someone willing to commit $5K to $50K to own fractional rights in a single Triceratops head with no clear exit liquidity for years. The team is assuming the first buyer set converts to the second. This is plausible but unproven, and the size of the gap will determine whether the platform’s economics actually compound.

The Oversubscription

The Labs raise opened on May 8 with a stated $200K target and a seven-day window. With approximately two days remaining as of May 12, the contract has received $9.87M in USDC commitments from 719 unique wallets across 934 individual deposits. That is roughly 49x oversubscribed.

The mechanical explanation of the 49x figure matters as much as the sentiment signal. Once a MetaDAO raise clears its capped amount, every additional dollar of deposit dilutes everyone’s allocation pro-rata, and the dilution gets worse as more capital piles in late. A wallet that committed $10K early in the round may end up receiving allocation as if it had committed a few hundred. Anyone who actually wants size on the launch comes out under-allocated relative to their target, and the natural response is to deposit more into the raise, which is what produces the feedback loop visible in most heavily oversubscribed MetaDAO launches.

What the oversubscription does tell you cleanly is that informed capital views the launch as positive EV. People committing six-figure deposits to a $200K-target raise know exactly how the mechanism works and what it does to their allocation. They are not bidding because they expect to receive their full check at launch. They are bidding because they expect the post-launch market to clear meaningfully higher than the launch price, and the small allocation they do receive is worth the cost of capital and refund risk.

After the Launch: What to Watch

The first SPV launch is the entire test. $TRCH1 is scheduled for Q3 2026. The team’s entire value proposition compresses into whether they can clear a clean first acquisition and launch. Every revenue line in the model assumes this works. The case for for $RAWR rests on the assumption that it does. If $TRCH1 stumbles, $RAWR does not have a fallback thesis.

Specific things worth tracking once $RAWR is trading and $TRCH1 is in progress:

• Specimen disclosure. The team has not publicly identified which Triceratops head they intend to acquire. Provenance documentation, bone count, authentication paperwork, and the third-party paleontological assessment are all things that should be public well before the SPV opens. Detailed and verifiable disclosure here would be a strong signal. Vague or late disclosure would be a meaningful flag.

• Buyer breadth on $TRCH1. A $1M SPV filled by 500 distinct wallets is a different signal than the same dollar amount filled by 20 wallets, most of whom are also $RAWR whales. The team needs to demonstrate that the SPV layer attracts buyers who specifically want fossil exposure, not just buyers rotating into the next thing the Labs entity launches. The breadth of the $TRCH1 buyer base is probably the cleanest measurable indicator of whether the team is genuinely creating a new asset category.

• Museum loan terms. The unit economics of the entire model assume custody costs are transferred to an accredited institution. The first signed loan agreement converts that assumption into a fact. The identity of the museum and the financial terms (who covers insurance, who covers conservation, who covers transit, what the display commitment looks like) will tell holders whether the operational thesis is workable or whether the team is going to need to absorb costs the model does not account for.

• $TRCH1 secondary trading. The FOSSIL buyback program and the BONE lending market are designed to provide synthetic liquidity until real secondary volume develops. The question is how quickly real two-way price discovery emerges. Healthy organic volume within the first 90 days would validate the model. A market that requires the Labs Strategic Reserve to function as the sole bid would compress the treasury faster than the model assumes and would call the medium-term economics into question.

• Cadence to SPV number two. The published target is 5 to 10 SPVs over 24 months. The second launch is when holders find out whether the platform is a repeatable originator or a one-off. SPV number two arriving quickly after $TRCH1 on improved terms would suggest the flywheel works. A nine-to-twelve-month gap would suggest the model is functional but slower than the bull case requires.

What You’re Underwriting

The framing in the launch materials has run ahead of the operating reality, and the qualitative case for the project rests almost entirely on things the team has not done yet. No fossil acquired, no museum signed, no SPV launched, no contracts deployed. The team’s background is content marketing and Solana community building, not alternative asset operations. The five revenue streams compress into a single bet on issuance cadence and on-chain buyer depth, with the FOSSIL program structurally active in exactly the conditions where it would hurt the balance sheet most.

What works is also real. The category is empty, the auction data supports the demand thesis at the top of the market, the museum loan model is sound in principle, and the MetaDAO mechanism caps the launch valuation while futarchy governs the treasury. The $516K FDV is consistent with a project that has identified a thesis and assembled a team to attempt it, rather than one that has demonstrated it can be executed.

The 49x oversubscription is doing the work people think it is doing, just for slightly different reasons than the narrative suggests. The structural feature of MetaDAO raises is that any wallet that wants real exposure has to over-deposit to defend against pro-rata dilution, which means the headline figure compounds on itself once a raise becomes visibly oversubscribed. What is signal in that number is that informed capital is willing to lock up six-figure deposits knowing most of it refunds, because they expect the post-launch market to clear meaningfully above the launch price. That is a reasonable read of the mechanism, not a guarantee of the business.

The right way to size $RAWR is as a small position priced for option value. The launch valuation is low enough that the team being right about the category and competent enough to clear the first SPV produces a real outcome. The launch valuation is low enough that being wrong about the team produces a livable outcome too. Where the position gets harder to defend is at any valuation that requires the issuance flywheel to have already turned. Above $3M to $5M FDV, the price is no longer pricing the option, it is pricing the outcome, and the outcome has not been demonstrated.

The first SPV will resolve most of the open questions. Until then, $RAWR is a wager on a thesis and on two people executing in a skill set they have not publicly demonstrated. The thesis has merit. The mechanism protects the entry. The team is the variable.

Alfreton Town F.C., a sixth-tier English football club in the National League North, launched a public token sale on April 21 through the AlphaFC platform. The sale is the first attempt to give football supporters real, legally enforceable governance rights over a professional club using Solana-based tokens. As of April 22, the raise has pulled in just over $75,000 USDC from 24 unique depositors across 35 transactions, with roughly four days remaining in the sale window.

This piece covers what AlphaFC is building, how the raise is structured, and what the on-chain data looks like so far.

What Is AlphaFC

AlphaFC is a platform for launching tokenized, fan-governed football clubs. Alfreton Town is the first club on the platform, but the model is explicitly designed for replication across many clubs. AlphaFC acquired Alfreton Town in June 2025, restructured it into a fan-ownership model, and secured sponsorships from Phantom, Raydium, and Bonk. The club’s stadium was renamed Solana Stadium.

The platform is not launching a meme coin. It is attempting to give token holders real, legally recognized membership in a UK nonprofit entity that owns and operates a professional football club. Every major decision the club makes, from hiring leadership to allocating budget to choosing kit designs, is voted on by token holders.

The Legal Structure

The club operates as a Company Limited by Guarantee (CLG), which is a standard UK nonprofit structure. Instead of shareholders with equity, a CLG has members. Each token acts as a pro rata unit of membership.

This is not an invented framework. Over 50 football clubs in the UK already operate as CLGs, including AFC Wimbledon at the EFL level. The difference here is that AlphaFC is using Solana tokens as the membership instrument, making ownership globally accessible and tradeable rather than locked behind geography and paperwork.

A BVI entity handles the token sale itself and distributes funds to the club through grants. The club’s board is constitutionally bound to respect token holder votes on major decisions, provided those decisions comply with league and government regulations. Token holders do not have a direct claim on profits or revenues, but all profits are reinvested into the club.

Token Allocation

The total supply is 100 million tokens, broken down as follows:

69% goes to the public sale, which is the raise happening now. 100% of funds raised go directly to club operations: player budget, facility upgrades, staffing, and community programs. AlphaFC takes no fee on this first raise.

10% is allocated to local fans. 1% is distributed immediately, and 2% per season is distributed over the next four seasons to supporters who attend matches and scan their Phantom wallet at the gate. This is designed to incentivize matchday attendance and reward the local community.

6% is held by the club treasury, ultimately governed by token holders, for marketing, player incentives, and other initiatives the community votes on.

Approximately 6.9% goes to liquidity, split across constant product and structured trading pools.

10% is retained by AlphaFC as an ecosystem fund. This allocation is intended to reward users of the broader platform over time. AlphaFC plans to eventually launch its own governance token, and holders of that token would receive allocations of club tokens from every club launched through the platform.

For future clubs, AlphaFC intends to take a percentage of both the raise amount and the token supply.

Club Economics

Football economics at the sixth tier are modest. Alfreton Town’s annual revenue profile breaks down roughly as follows.

Gate receipts are typically the largest source for clubs at this level. Something on the order of 500 to 600 fans per home match at around $13.50 per ticket across 23 home matches.

Sponsorship is where the crypto-native positioning has already paid off. Phantom, Raydium, and Bonk have contributed approximately $367,000 in sponsorship revenue. Including pitchside billboards and other placements, total sponsorship rounds up to approximately $405,000. This is an unusually large figure for a club at this tier and is a direct result of positioning as Solana’s football club.

Competition prize money offers meaningful upside. The FA Cup, the oldest and most prestigious cup in English football, is open to clubs all the way down to the eighth tier. If Alfreton wins through the qualifying rounds and reaches the first round proper, a single match can generate $135,000 to $270,000, especially if drawn against a Premier League side where gate revenue is split between the two clubs.

TV and media distributions from the National League’s enterprise sponsorship amount to roughly $67,500 to $135,000 per year.

The core thesis AlphaFC is betting on: most clubs at this level are structurally unprofitable and depend on wealthy individuals treating them as passion projects or tax write-offs. Crypto-native capital formation, global fan engagement, and tokenized governance change that equation by bringing in supporters and sponsors from around the world.

The Raise So Far: On-Chain Data

The token sale launched on April 21 and the deposit address has received 35 inbound USDC transfers from 24 unique wallets, totaling $75,118 USDC as of April 22 at 09:22 UTC. There are approximately four days left in the raise.

The bulk of the capital arrived quickly. Within the first two hours after launch, $64,929 had already been deposited, representing 86% of the current total. Activity has tapered significantly since, with the most recent hours seeing only single-digit dollar deposits trickling in.

Breaking it down by size bucket: two deposits over $10,000 account for $31,994. Five deposits between $5,000 and $10,000 account for $35,341. Three deposits between $1,000 and $5,000 make up $3,965. Seven deposits between $100 and $500 total $1,621. And the remaining deposits under $100 combined contribute $691.

Executive Summary

Meteora closed Q1 2026 with $11.4M in protocol revenue, $6.8M in operating profit, and a $32.8M treasury. Revenue declined 35% QoQ but the protocol swung from Q4’s -$0.8M operating loss to profitability, driven by two structural changes: the elimination of fee token depreciation and a sharp normalization in opex.

The revenue mix tells the real story. Partner fees doubled to $6.6M on a January Bags.app surge, but LP pool revenue, the recurring foundation, fell 65% to $4.8M, its weakest quarter on record. By March, total monthly revenue had fallen to $1.3M. Cash generation was the bright spot: $25.4M in inflows and $18.3M in net cash flow, boosted by the new auto-conversion mechanism sweeping previously uncollectable fees into the treasury.

Competitively, Meteora’s headline revenue decline of 35% landed between Orca (-43%) and Raydium (-24%), but stripping out the launchpad windfall, the core LP business underperformed both peers. The prop AMM share shift continued, and Pump.fun generated 18x more launchpad revenue than Meteora’s DBC partners. The Dynamic Terminal, Bedrock, and upcoming Native Limit Orders are the strategic responses, but their revenue impact is a 2026 story yet to be written.

Income Statement

Q1 2026 saw Meteora return to profitability after Q4’s operating loss, posting $6.8M in operating profit against $11.4M in protocol revenue.

On the revenue side, the picture inverted from Q4. Partner/launchpad fees were the standout at $6.6M, up 86% QoQ, driven almost entirely by a January surge through Bags.app that generated $5.1M in a single month, outpacing Q4’s entire partner revenue of $3.5M. February dropped to $1.1M and March to $371K as launch activity subsided. Post-bonding trading fees continued to dominate at $5.2M versus $1.3M from pre-bonding, reinforcing the FY2025 pattern that launchpad revenue comes primarily from sustained secondary trading rather than initial bonding curve participation.

The core LP pool business told the opposite story. DLMM revenue fell to $3.1M, down 61% QoQ from Q4’s $8.1M, as the broader Solana DEX market contracted and lower volatility reduced the dynamic fee mechanism’s upcharge frequency. DAMM came in at $1.6M, down 71% QoQ from $5.7M, with the steeper decline partly reflecting migration of newer pool creation toward DAMM v2 and DBC, which are categorized under partner revenue. Total LP pool revenue of $4.8M was the weakest quarter since Meteora began reporting financials.

Fee token depreciation was zero for the first time. The automatic fee-to-USDC conversion mechanism deployed in January has effectively eliminated this line item by sweeping long-tail token fees into stablecoins at the program level. In Q4, the estimated $6.9M depreciation charge was the single largest drag on the income statement. Its removal is the most meaningful structural improvement to Meteora’s economics since the protocol began publishing financials, and it’s one large reason Q1 shows a $6.8M profit despite revenue declining 35%.

Operational expenses dropped sharply to $4.2M from Q4’s $10.7M, falling below the FY2025 quarterly average of roughly $4.7M. Capital expenditures were minimal at $388K. Together, total expenses of $4.5M were the lowest since Q2 2025’s $1.6M (when the team was still scaling).

Cashflow Statement

The cashflow statement in Q1 tells a more encouraging story than the income statement, with cash generation outpacing reported revenue by a wide margin. Operating cash flow came in at $21.2M, nearly double the $11.4M in reported revenue. The disconnect is driven by the auto-conversion mechanism deployed in January, which is now systematically sweeping long-tail token fees accumulated across thousands of pools into USDC. Q1 cash inflows of $25.4M, up 30% from Q4’s $19.5M, include a catch-up effect as previously uncollectable fees earned in prior quarters were converted for the first time. This divergence should narrow in future quarters as the backlog clears and cash inflows converge with current-period revenue.

MET-related outflows totaled $2.5M, down sharply from Q4’s $19.4M TGE-driven spike. Of this, $1.0M went to buybacks executed through market makers, acquiring 7.0M MET at an average price of $0.1427. The remaining $1.5M was lost to an OTC scammer during an attempted token buyback. The team disclosed this directly and has filed a police report. Cumulative buybacks now stand at $13.7M, with 39.6M MET acquired representing 3.97% of total supply.

Capital expenditure was minimal at $388K. Combined with $4.2M in operational spend, total outflows of $7.0M were a fraction of Q4’s $30.8M, which was inflated by non-recurring TGE costs. Net change in cash was +$18.3M, a meaningful reversal from Q4’s -$11.3M outflow, bringing the quarter-end treasury to $32.8M.

Balance Sheet

Meteora’s treasury at the end of Q1 stood at $32.8M, down from $35.9M at year-end despite the quarter generating $18.3M in positive net cash flow. The disconnect is expained by a drop in the value of nonstable assets on the balance sheet: the treasury held a significant non-stable position throughout the quarter, and as the price declined, the mark-to-market value of those holdings fell even as new cash was flowing in. The treasury composition shifted toward stablecoins, with USDC/USDT more than doubling from $6.9M to $15.1M while SOL/WSOL fell from $22.2M to $11.7M. An additional $6.0M sits in other token holdings. This rebalancing toward stables is prudent treasury management for a protocol that funds operations in dollar terms.

At Q1’s operational spend of $4.2M per quarter, the $32.8M treasury represents roughly 7.8 quarters, or just under two years, before accounting for any fee income. That’s a meaningful improvement from the runway picture at year-end, where Q4’s $10.7M run rate implied only 3.4 quarters of coverage. The Q1 opex figure coming in below the FY2025 quarterly average suggests Q4’s elevated spend was at least partially front-loaded. More importantly, the protocol generated $21.2M in operating cash flow this quarter even in a soft market, making Meteora comfortably self-sustaining.

Protocol KPIs

Q1 volumes declined across the board but the launchpad segment bucked the trend. DLMM volume came in at $17.0B, down 39% QoQ, while DAMM pools did $1.8B, down 23%. DBC was the outlier, nearly tripling to $718M from Q4’s $244M, driven by the January Bags.app surge. Total pool TVL actually recovered to $782.8M from Q4’s $716.0M, a modest bright spot against the broader volume decline.

LP fees told the more important story. Launchpad-related pool fees nearly doubled to $34.2M from $14.2M in Q4, again a January phenomenon. Other pool fees fell sharply to $71.8M from Q4’s $187.5M, a 62% decline that tracks the combination of lower volumes and compressed take rates in a low-volatility environment.

On the DBC activity side, token launches rose to 77,385 from Q4’s 42,057, and graduated tokens jumped to 1,087 from Q4’s 225. The graduation rate improved to 1.4% from Q4’s 0.5%. The increase was driven by the same Bags.app wave that lifted partner revenue in January rather than any structural improvement in launch quality. As with partner fees, the monthly trajectory within Q1 likely showed heavy front-loading with a sharp fade through February and March.

Ecosystem KPIs

Q1 saw crypto market cap fall another 24% from $2.98T to $2.28T, with Solana DEX volume declining 17% to $284.5B. Unlike Q4, where Solana volumes held flat against a falling market, this quarter saw on-chain activity contract in line with broader valuations.

Revenue declines were steeper than volume declines across Solana’s major DEX protocols, consistent with the fee compression trend identified in Q4. Meteora’s protocol revenue fell 35% QoQ to $11.4M. Raydium dropped 24% to $5.6M, while Orca fell 43% to $1.8M. Meteora’s relative performance was middle of the pack again, though the comparison is somewhat distorted by the January launchpad burst that inflated its top line. Stripping out partner revenue, Meteora’s core LP revenue of $4.8M represents a 65% decline, worse than both peers and a signal that the core liquidity business felt the downturn more acutely than the headline number suggests.

On the bonding curve side, Pump.fun revenue came in at $119.8M, actually up 11% QoQ from Q4’s $107.8M, while Meteora’s DBC partner fees rose 86% to $6.6M. Both platforms benefited from the January launch wave, but the gap in scale remains stark: Pump.fun generated 18x more revenue than Meteora’s launchpad segment. Pump.fun continues to operate as the default high-volume launchpad on Solana, with a consistency that Meteora’s partner-dependent model hasn’t matched. That said, Q1 was a better quarter for DBC than Q4, with the Bags.app surge demonstrating that when a partner catches momentum, the revenue contribution can be meaningful. The challenge remains the same: sustaining that momentum beyond individual platform cycles.

The prop AMM trend flagged in Q4 continued through Q1, with specialized vertically integrated liquidity solutions maintaining volume share at the expense of traditional DEXs. Meteora’s Dynamic Terminal launch and upcoming Native Limit Orders are direct responses to this competitive pressure, aiming to capture trader flow that might otherwise migrate to purpose-built AMMs. Whether these product initiatives can reverse the structural share loss remains the key open question heading into Q2.

Qualitative analysis

Q1’s focus shifted from building infrastructure to expanding who uses it. The Dynamic Terminal was the headline launch, a trading-native UI designed to let users combine buying, selling, and LP strategies in one interface, targeting the broader on-chain trader population beyond dedicated LPers. Native Limit Orders were teased as coming soon at the April 9 LP Roundtable.

Bedrock launched in late March as a joint venture with GVRN, providing legal infrastructure that gives token holders enforceable ownership rights via ring-fenced SPVs. It’s already integrated with Bags.app, Star.fun, America.fun, Collaterize, and Fair.Club. If it attracts founders who want real equity backing behind their tokens, it could drive a more durable launchpad revenue stream less dependent on memecoin cycles.

PreStocks emerged as an unexpected growth area, with tokenized pre-IPO volumes on Meteora growing from $22M in January to $207M in March across names like Anthropic, SpaceX, and xAI. At 4.5% of total volume by quarter-end it’s still small, but it’s genuinely new demand that didn’t exist a year ago.

Executive Summary

Collector Crypt posted its strongest quarter on record in Q1 2026, with estimated total revenue of $146.9M and gross profit of $8.6M, both all-time highs. Revenue remained overwhelmingly driven by the Gachapon product, which generated $144.7M of the quarter’s top line. The quarter was also notable for diversification within the Gachapon itself, as the $1K Pokemon pack became the top revenue-generating pack and a $250 One Piece pack crossed $10M in volume, broadening the product mix beyond the $50 and $250 Pokemon packs that dominated prior quarters.

Gross margins, however, remain structurally compressed. After operating in the 10-12% range through Q3 2025, margins fell to 5.7% in Q4 2025 and recovered only slightly to 5.9% this quarter. Revenue concentration among high-volume users also continued to intensify, making top-line growth increasingly dependent on a narrow cohort of heavy spenders.

The treasury grew 7.3% quarter-over-quarter to $18.6M, with Pokemon cards accounting for 81% of holdings. One Piece emerged as the fastest-growing treasury segment, reflecting the category’s momentum on the Gachapon side.

In the broader competitive landscape, Collector Crypt outpaced both the card grading market (which grew 8.9% QoQ) and smaller competitor Phygitals (which declined 71%), though market leader Courtyard widened its gross profit lead. Renewed mainstream interest in Pokemon cards, reflected in near-peak Google search trends, provided a favorable demand backdrop but also raises questions about the sustainability of current growth rates if that attention fades.

Income statement

Q1 2026 was Collector Crypt’s strongest quarter on record, with total revenue reaching an estimated $146.9M and gross profit hitting $8.6M. Both figures represent all-time highs, continuing a streak of sequential quarterly growth that has held since the protocol’s launch in Q4 2024. Revenue composition remained heavily concentrated in the Gachapon product, which accounted for $144.7M of the quarter’s top line.

Redemption/Burn Fees, Marketplace Royalties, and eBay Sniper Fees all posted their largest quarterly figures to date, though these line items are estimated and remain marginal relative to Gachapon volume.

Gachapon gross margin tells a more nuanced story. The protocol operated at roughly 10-12% gross margins through its first three full quarters (Q1-Q3 2025), with pack expenses tracking revenue at a relatively stable ratio. Q4 2025 marked a sharp compression to 5.7% as pack costs grew faster than revenue. Q1 2026 saw a modest recovery to 5.9%, but the margin profile has clearly shifted structurally lower compared to the sub-$100M revenue era. While the protocol is generating more gross profit in absolute dollars than ever before, it is doing so on significantly thinner margins, suggesting the revenue mix at scale carries a higher cost basis than earlier, lower-volume quarters.

One thing that characterized this quarter, along with record revenue and gross profit, was the number of new pack types gaining traction. For most of Q4 2025, $50 and $250 Pokemon packs represented nearly all the volume. This quarter, the $1K Pokemon pack became the most popular pack in terms of revenue, and a $250 One Piece pack had over $10M in volume.

One notable trend that continued this quarter is the growing concentration of revenue among large users on the app. Despite steady growth in the number of wallets using Collector Crypt’s Gachapon, the upper end of these users is becoming more pronounced, driving an increasing share of marginal volume. This makes revenue on the app more fragile and dependent on these high rollers.

Treasury

Collector Crypt’s cardtreasury has grown substantially alongside the protocol’s revenue expansion. Total treasury assets reached $18.6M by the end of Q1 2026, up from $17.3M at the close of Q4 2025—a 7.3% quarter-over-quarter increase.

The composition of the treasury remains heavily weighted toward Pokemon cards, which account for $15.1M or 81% of total holdings. This concentration reflects both the dominance of Pokemon packs in Gachapon volume and the protocol’s strategy of accumulating inventory in its highest-demand category.

One Piece has emerged as the fastest-growing treasury segment. From virtually zero at the start of Q3 2025, One Piece holdings have grown to $1.56M, driven by the successful launch of the $250 One Piece pack this quarter. Sports cards also saw meaningful growth, reaching $713K as the $100 Sports pack gained traction.

Meanwhile, legacy positions in MTG and Yugioh have remained relatively flat or declined, suggesting these categories may be deprioritized as the protocol focuses inventory acquisition on higher-velocity products.

Performance in Context

Collector Crypt’s Q1 2026 results can be evaluated against several ecosystem KPIs that provide context for the protocol’s growth trajectory.

Competitive Landscape

Courtyard, the leading tokenized collectibles platform, generated an estimated $13.2M in gross profit during Q1 2026, up 48% from $8.9M in Q4 2025. Collector Crypt’s $8.6M gross profit grew 30% over the same period—solid growth, but notably slower than Courtyard’s pace. The gap between the two platforms widened in absolute terms, with Courtyard now generating $4.6M more in quarterly gross profit compared to $2.3M in Q4 2025.

Phygitals tells a different story. After peaking at $4.9M in gross profit during Q3 2025, the platform declined to $3.6M in Q4 2025 and collapsed to just $1.0M in Q1 2026—a 71% quarter-over-quarter drop. This suggests the onchain collectibles market may be consolidating around the top players, with Courtyard pulling further ahead while smaller competitors fall behind.

Grading Market Activity

Total cards graded industry-wide reached 5.96M an ATH in Q1 2026, up 8.9% quarter-over-quarter from 5.47M in Q4 2025. This growth rate has moderated from Q4’s 11.7% pace but remains healthy. For context, Collector Crypt’s Gachapon revenue grew 25.6% over the same period—nearly 3x the rate of the underlying grading market—suggesting the protocol is capturing share rather than simply riding industry tailwinds.

Consumer Interest

Google search trends for “Pokemon cards” reached 95 in Q1 2026, the highest quarterly reading since Q1 2025 (88) and well above the Q2-Q3 2025 trough (55-58). This renewed search interest correlates with Collector Crypt’s record Gachapon volume and suggests the protocol benefits directly from mainstream Pokemon enthusiasm.

Our Thoughts

Uncorrelated Growth

a major encouraging signal is the consistency of Collector Crypt’s growth despite large fluctuations in the broader crypto market. While most onchain applications see activity rise and fall in lockstep with token prices and market sentiment, Collector Crypt’s transaction volume has shown remarkably low correlation to crypto market conditions. This suggests the protocol is tapping into genuine consumer demand for collectibles rather than speculative crypto-native behavior—a meaningful differentiator that could provide stability through future market cycles.

Competitive Positioning

Despite this very positive quarter, Collector Crypt is losing ground to Courtyard. As noted in the Performance in Context section, Collector Crypt was ahead of Courtyard just two quarters ago—now it trails by $4.6M in quarterly gross profit. Our thinking is that to gain ground here, Collector Crypt must build out more retail-friendly ways for people to use their app, like the mobile app that Courtyard has, as nearly none of Collector Crypt’s volume comes from non-stablecoin payments, meaning that the platform is not as accessible to non-crypto natives.

Revenue Concentration Risk

Perhaps the most pressing concern heading into Q2 2026 is the protocol’s increasing dependence on a small cohort of high-volume users. As the Average Spending Per User chart illustrates, monthly spending per wallet has more than doubled over the past year—from roughly $3.5K in mid-2025 to $8.8K in March 2026. While this could reflect deeper user engagement, it more likely indicates that a shrinking pool of whales is driving an outsized share of Gachapon volume.

This concentration creates fragility. If even a handful of top spenders reduce activity—whether due to market conditions, competitive alternatives, or simple fatigue—the impact on revenue could be severe. The protocol’s record top-line numbers are impressive, but they rest on an increasingly narrow foundation. Diversifying the user base and reducing whale dependency should be a strategic priority.

Executive Summary

Q1 2026 was a consolidation quarter for Zcash. After the extreme attention and outperformance of Q4 2025, most network metrics cooled: hashrate declined slightly, ZEC underperformed BTC on price, transaction volume fell from Q4 highs, and external attention bled off from the Q4 spike. However, all of these figures remained starkly above Q3 2025 levels, confirming that last quarter’s momentum left a durable imprint on the network rather than a fleeting one.

The critical exception was shielded pool growth. The amount of ZEC held in shielded pools continued climbing throughout Q1, with net inflows outpacing new issuance by roughly 76%. This means the growth was driven by existing holders migrating transparent balances into shielded pools, not simply by the accumulation of block rewards. Orchard absorbed nearly all of the increase, adding roughly 300K ZEC while Sapling and Sprout continued to decline. For a network whose value proposition rests on private value storage, sustained shielded adoption through a quieter market period is the most important signal in the report.

On the development side, ZODL, the newly independent core engineering team, closed a $25M+ seed round from Paradigm, a16z crypto, Coinbase Ventures, and others while shipping multiple wallet releases. The Zcash Foundation delivered three Zebra node releases including critical security patches and advanced NU7 governance polling. Shielded Labs progressed Crosslink toward productionization with testnet workshops and an independent security review. Project Tachyon secured near-universal community support for its scalability overhaul during formal sentiment polling. All four organizations coordinated effectively on a critical Sprout pool vulnerability in late March, delivering a fix within 24 hours and deploying it across major mining pools within three days with no funds lost.

Overall, Q1 is best understood as a quarter of renewed long-term participation, institutional credibility, and consistent growth in the chain’s most important KPI, layered on top of a natural cooldown from Q4’s enthusiasm.

Shielded pools

Zcash distributes its total coin supply across five distinct pools that each serve a different function: Transparent holds ZEC that behaves like Bitcoin, fully visible on-chain; Sprout and Sapling are legacy shielded pools that use progressively stronger zero-knowledge cryptography to hide transaction details; Orchard is the current-generation shielded pool offering the strongest privacy guarantees and lowest transaction costs; and Lockbox is a governance-controlled reserve that accumulates a share of block rewards for community-directed spending. The distribution of supply across these pools is the single most important lens for evaluating Zcash's network health because it reveals whether users are actually adopting the privacy features that justify ZEC's existence as a distinct asset. A network where 95% of supply sits in Transparent is functionally just a smaller Bitcoin. Migration from Transparent into shielded pools, and specifically from legacy pools into Orchard, signals genuine demand for privacy and growing sophistication among holders, while the relative size and volume in/out of each shielded pool reflects the strength of the anonymity set that protects all participants. Tracking these numbers over time tells you whether Zcash is delivering on its core thesis or just trading on narrative.

Pool balances

In Q1 2026, the amount of ZEC in shielded pools grew by approximately 5.9%, representing roughly 288K new ZEC deposited into these pools on a net basis. This growth is about 76% larger than the amount of new ZEC issued through mining this quarter, showing that shielded pool inflows are outpacing new token issuance, although the rate of growth has slowed relative to the dramatic expansion in shielded pool supply seen last quarter. Breaking down the growth by pool, the Orchard pool grew by nearly 300K ZEC over the quarter while the Sapling pool declined by about 12K ZEC and the Sprout pool declined by roughly 50 ZEC. In terms of the overall composition of shielded ZEC, this quarter represented a slower continuation of last quarter's pool growth and a continued concentration into Orchard as the dominant shielded pool.

Pool flows

Throughout Q1 2026, the USD value of ZEC flowing in and out of shielded pools dropped by 50-60% relative to the start of the quarter. Roughly half of this decline can be attributed to the drop in ZEC’s price, while the remainder reflects a genuine reduction in the volume of ZEC moving through shielded pools. When breaking down flows by pool, the dynamic diverges meaningfully: the Orchard pool experienced a much sharper drop in flow activity than Sapling on a percentage basis. In contrast, the decline in Sapling pool flows can be more than fully explained by the price drop alone, as the pool actually saw more ZEC moving in and out on a native token basis than it did in Q4 2025.

Network demand metrics

The metrics in this section measure the external demand side of the Zcash network: how much value is moving across the chain, how much users are paying to move it, and how much attention the network is attracting from outside the existing holder base. Where the shielded pools section above measures the depth of commitment among current participants, these metrics capture the breadth of activity and whether new usage is flowing into the network. In Q1, the broad pattern across all demand metrics was a gradual decline from Q4 peaks that was largely explained by ZEC's price drop rather than a meaningful reduction in native token-denominated activity. The network is quieter than it was three months ago, but the underlying usage measured in ZEC has held up better than the USD-denominated figures suggest.

Transfer Volume

During Q1 of this year, total transfer volume on the Zcash network has generally been on a downtrend, with a spike at the end of February. This downtrend is nearly entirely due to the price decline of ZEC, as transfer volume denominated in ZEC is comparable to the end of Q4 2025.

Network fees

Network fees tell the same story as transaction volume, declining through the quarter mostly as a result of declining ZEC price. Notably, transaction fees on the network are only a few thousand dollars a week, so similar to Bitcoin, the vast majority of revenue received by miners comes from block subsidies and there are no significant fee markets on the network that can independently fund miners.

Attention

When looking at Google search trends and other metrics that represent external attention on the network, this quarter has shown a consistent bleed in attention from the massive spike the network received during its price pump in Q4 of 2025.

Mining metrics

In Q4 2025, the hashrate of Zcash increased by about 75%, while Q1 saw a slight cooling off from that strong run. When looking at the number of mining pools winning blocks per day on the network, that figure has increased from about 12 per day at the start of the quarter to 14-16 per day currently. In July 2025, there were only 7 pools winning blocks, so miner diversity has roughly doubled. Given that mining difficulty held up better than ZEC's price this quarter and pool diversity continued to increase, the modest cooling in hashrate actually reflects growing long-term commitment to the network's infrastructure rather than a retreat by miners.

Network development

Zcash development is driven by four primary organizations, each with a distinct mandate, with development funding flowing through a fifth entity, Zcash Community Grants. ZODL (Zcash Open Development Lab), the original engineering team behind Zcash, leads product development including the Zodl mobile wallet and third-party SDKs, while also serving as the network’s primary marketing and communications arm. The Zcash Foundation operates as an independent nonprofit focused on privacy as a public good, maintaining the Zebra node implementation to ensure client diversity and organizing community events such as Zcon. Tachyon is a specialized engineering team dedicated to long-term scalability, working on an architectural overhaul that would use recursive zero-knowledge proofs to reduce storage and bandwidth requirements and allow the network to handle significantly higher transaction throughput. Shielded Labs functions as a third-party research team currently focused on Crosslink, a proposed hybrid Proof of Work and Proof of Stake consensus mechanism. Zcash Community Grants allocates a portion of block rewards based on tokenholder voting, a governance structure formalized in Q4 2025 that replaced earlier, more centralized funding models.

ZODL

ZODL (Zcash Open Development Lab) was formed in January 2026 after the entire core engineering and product team resigned from Electric Coin Company following a governance dispute with Bootstrap, the nonprofit board overseeing ECC. The team took the rights to the flagship Zcash mobile wallet (originally Zashi, rebranded to Zodl) and established ZODL as an independent entity.

In March, ZODL closed a seed round exceeding $25 million from Paradigm, a16z crypto, Coinbase Ventures, Winklevoss Capital, Maelstrom (Arthur Hayes), Cypherpunk Technologies ($5M), Chapter One, and angels including Balaji Srinivasan and Haseeb Qureshi. Funds are earmarked for team expansion, wallet development, and core ZEC protocol work.

On the product side, ZODL shipped Zodl wallet versions 3.2.0 through 3.3.1 during the quarter, adding Keystone hardware wallet support, SDK upgrades, and UX improvements. The wallet reached roughly 40k unique iOS installs and a 4.9-star App Store rating. Broader ecosystem metrics tied to wallet adoption include a 400%+ increase in the Orchard shielded pool and over $600M in shielded ZEC swaps since late 2025.

ZODL also played a central role in remediating a critical security vulnerability discovered in late March. A white-hat researcher found a flaw in zcashd node software, present since July 2020, that could have allowed nodes to skip proof verification for the deprecated Sprout shielded pool. In a worst case, a malicious miner could have drained roughly 25,000 ZEC ($6-7M) from the pool. ZODL engineers authored a fix within 24 hours, major mining pools deployed the patch within three days, and the fix shipped in zcashd v6.12.0 on March 31. No funds were lost and the bug was never exploited. The response was widely viewed as a strong signal of operational credibility for the newly independent team.

Zcash Foundation

The Zcash Foundation (ZF) continued operating as the primary maintainer of the Zebra Rust node implementation and played a central stewardship role in governance, events, and ecosystem funding throughout Q1.

On the engineering side, ZF shipped three Zebra point releases. Zebra 4.1.0 (February 9) expanded observability with Prometheus metrics. Zebra 4.2.0 (March 17) added RPC functionality and stronger mempool policy compliance. Zebra 4.3.0 (March 26) included critical security fixes addressing two high-severity issues (a consensus bug and a remote DoS/panic vector), ZIP-235/NSM support, and improved profiling tooling. This release cadence supports the long-term goal of making the Z3 stack (Zebra + Zaino indexer + Zallet) the primary production node infrastructure, particularly as the ecosystem moves toward deprecating the legacy zcashd node after Network Upgrade 7. ZF also continued advancing FROST (Flexible Round-Optimized Schnorr Threshold signatures), including progress on COCKTAIL-DKG tracking and preparation for v3.0.0 and ZIP-312.

On the governance and community front, ZF ran multiple rounds of ZCAP and coinholder sentiment polling for NU7 ZIPs in February, publishing a summary (”What We Heard and Where We Go From Here”) on February 23. In March, they hosted Zcomm 2026, a one-day virtual community-builders event focused on practical skills for running hackathons, building partnerships, and designing privacy-focused communities. ZF also completed a full cycle of the Coinholder-Directed Retroactive Grants Program during Q1, from the call for proposals in January through coinholder polling in March, and continued administering Zcash Community Grants including regional ambassador and education programs.

On the regulatory front, the long-running SEC investigation into the Foundation was closed without enforcement action in January 2026. ZF also participated in the coordinated response to the Sprout verification bug in late March, splitting the 200 ZEC white-hat bounty alongside ZODL, Shielded Labs, and Bootstrap.

Shielded Labs

Shielded Labs, a Swiss-based, donation-funded nonprofit, had a quietly productive Q1 focused on long-term protocol sustainability, finality research, and governance stewardship rather than user-facing products.

On the funding side, Shielded Labs received two notable donations reinforcing its independence from block reward-based funding. The Winklevoss twins contributed approximately $1.2 million (3,221 ZEC) in January, their second donation to the organization. Vitalik Buterin followed in February with a donation targeted specifically at Crosslink development, also his second contribution to Shielded Labs.

Crosslink remained the organization’s flagship project: a proposed hybrid PoW + PoS finality layer designed to add faster finality, improved 51% attack resistance, safer bridges, and the potential for staked ZEC to become a yield-bearing asset. Q1 progress included continuation of Milestone 4 work, a Crosslink workshop (Milestone 4C) held January 28 in collaboration with the Zingo! team where participants tested staking actions and finalizer UI/UX on testnet, and completion of an independent security and mechanism-design review (led by Nikete) that validated the finality design and identified potential improvements. The team aimed for a working testnet prototype by early Q1 with productionization targeted later in 2026.

Shielded Labs played a central coordinating role in the Sprout verification bug incident in late March. The white-hat researcher reported the vulnerability directly to Shielded Labs on March 23. The following day, Shielded Labs convened with ZODL engineers to validate the issue and develop a fix, then coordinated with major mining pools for rapid deployment. Shielded Labs co-published the joint security disclosure and contributed 50 ZEC toward the 200 ZEC bounty.

On governance, Shielded Labs participated in the Coinholder-Directed Retroactive Grants Program as one of the Key-Holder Organizations alongside ZF, exercising veto rights on certain proposals per ZIP 1016 rules. They also contributed to NU7 polling discussions and publicly urged the community in early March to act quickly on long-term sustainability changes (NSM, dynamic fees, issuance smoothing) to attract patient capital.

Tachyon

Project Tachyon is a community-led, independent protocol upgrade initiative proposed in April 2025 by cryptographer Sean Bowe, focused on re-architecting shielded transaction handling to achieve massive scalability and post-quantum privacy without creating a new shielded pool.

Q1 2026 was a consensus-building and visibility quarter rather than a shipping quarter. No mainnet activation or major code release occurred, with full migration expected later in 2026. In January, Sean Bowe appeared on the Zero Knowledge Podcast for a detailed interview on Tachyon’s design and roadmap, raising visibility right after the ecosystem restructuring. In February, Tachyon received near-universal support across every polled group (core engineers, ZCAP, coinholders) during the Zcash Engineering Caucus protocol feature sentiment polling, with roughly 30 support votes versus 4-5 oppose on key questions. The Zcash Foundation highlighted this backing in their NU7 polling results summary published February 23, directly influencing the NU7 direction. In March, a fully dedicated Zcash Engineering Office Hours session (March 17) covered technical progress on the Ragu PCD crate and next steps. ZODL also explicitly referenced ongoing Tachyon work as part of their 2026 roadmap planning and investor discussions.

Tachyon remains community and donation-funded with no major grants or funding announcements specific to the project in Q1. As of early April 2026, major scalability improvements are on track for roughly one year after the original proposal, with full migration expected later.

Executive Summary

What is Hylo?

Hylo is a protocol on Solana that creates leveraged exposure to SOL with liquidation resistance. The protocol has three core components:

• xSOL — A leveraged SOL token that amplifies price movements. When SOL goes up, xSOL goes up more; when SOL goes down, xSOL goes down more. Unlike traditional leverage, positions are liquidation resistant.

• hyUSD — A USD-pegged stablecoin backed by liquid staked SOL collateral (currently JitoSOL and HyloSOL). It maintains its peg through arbitrage incentives and the stability pool mechanism.

• sHYUSD — A yield-bearing version of hyUSD. Holders earn yield generated from the staked SOL’s staking rewards. Under normal conditions, sHYUSD maintains a stable value while generating returns.

The system works through a balancing act: xSOL holders take on leveraged SOL exposure, while hyUSD/sHYUSD holders take the other side, earning yield for providing stability. When markets move sharply, the stability pool can activate—using sHYUSD backing to purchase xSOL and defend the hyUSD peg.

Q1 2026: The Quarter in Review

Q1 2026 tested Hylo’s risk architecture under real market pressure. Protocol TVL fell 58% from $52M to $22.8M as SOL dropped 33%, triggering multiple stability pool activations that deployed $7.9M in hyUSD to defend the peg. These activations converted stability pool deposits into xSOL, fundamentally shifting sHYUSD from pure stablecoin yield to a hybrid position with leveraged SOL exposure. The result: sHYUSD delivered its strongest quarter on record at 52.4% APY as xSOL recovered from lows, validating the protocol’s volatility absorption mechanism.

Revenue performance remained structurally sound despite the protocol TVL decline. The protocol generated $434K in fees, representing 1.8% of quarter-end TVL, which was a very similar rate to previous quarters.

Late March brought Hylo’s most significant product evolution with the V2 beta launch at v2.hylo.so/early-access. The redesigned interface introduced streamlined trading flows, real-time collateral ratio monitoring, and improved risk transparency across the dual-token system.

KPIs

sHYUSD and the Stability Pool

The way Hylo’s stability pool works is that when the collateralization of hyUSD gets too low, as defined by the protocol, Hylo starts actively burning the hyUSD that backs sHYUSD and replacing that backing with xSOL, so the total supply of hyUSD stays within healthy ranges of the protocol’s collateral. Then, as the protocol’s numbers normalize, the stability pool that backs sHYUSD slowly sells off its xSOL to return to full stablecoin backing. The effect of these activations is that between the time of activation and the full sell-off of xSOL, sHYUSD is no longer a stable asset and fluctuates based on the price of SOL. Unintuitively, this quarter the exposure to xSOL that sHYUSD has had actually turned a profit for holders and made it the best-performing backing asset.

TVL Decline

This quarter saw a major hit to Hylo’s Protocol TVL, which dropped 58%. This decline came from two factors: (1) the price of the protocol’s collateral declining and (2) net withdrawals of collateral from the platform. Of this decline, 33% can be attributed to collateral dropping in value, but the remaining 25% ($13M) is a result of users pulling assets from the protocol.

The core dynamic at play is structural: for Hylo to maintain its stablecoin TVL while SOL is declining, new capital must enter the system willing to take on leveraged SOL exposure via xSOL. In Q4 2025, this is exactly what happened, and the protocol absorbed SOL’s 40% decline without triggering stability mechanisms. Q1 broke that pattern. As SOL continued to fall, demand for leveraged long exposure dried up, and without fresh xSOL capital to absorb the collateral decline, the protocol’s ratios drifted toward undercollateralization. This is what forced the stability pool activations described above, reducing the stablecoin supply to bring it back in line with available collateral.

The two cohorts holding money in the protocol were affected differently. xSOL holders had obvious reasons to exit during a sustained SOL decline. sHYUSD holders wanted stable yield, but the volatility profile introduced by the stability pool activations made the asset less appropriate for looping strategies and less comfortable to hold. Both cohorts pulling back simultaneously is what drove the $13M in net withdrawals beyond what price alone would explain.

Protocol Revenue

Hylo’s revenue comes from three major sources: minting/redemption fees on xSOL, a cut from the yield that would otherwise go to sHYUSD holders, and fees associated with their LST and validator. For the past two quarters, Hylo’s revenue from these sources combined has been about 1.8% of their quarter-end Protocol TVL, compared to about 1.3% in the two quarters before that. This quarter, the revenue drop was in line with the Protocol TVL decline. When viewing Hylo as an asset manager, the current fees they are accruing represent a very favorable carry at 5.2% to 7.2% of assets under management per year.

Ecosystem Context

Hylo underperformed the broader stablecoin market this quarter, largely as a delayed consequence of its Q4 resilience. The protocol absorbed SOL's decline last quarter by attracting new xSOL buyers, but this meant the system was quietly building up leverage even as the broader market was already compressing. When Q1 broke the pattern and fresh leverage demand dried up, the correction arrived all at once and hit harder than it otherwise would have, precisely because the leverage had been accumulating through a deteriorating environment rather than unwinding alongside it. Protocols like Ethena and other crypto-dependent yield platforms took their hits more gradually in Q4 and have either stabilized or seen slight growth this quarter, while Hylo's TVL decline was concentrated into a single quarter. This divergence in timing, rather than any structural weakness unique to Hylo, explains most of the relative underperformance.

Product Development: Hylo v2

While navigating market turbulence, the team continued building. Hylo v2 early access launched in late March.

What’s New in V2

The v2 overhaul addresses the most common friction points users encountered in v1.

On the trading and position management side, the update streamlines the stake/unstake flow for sHYUSD conversions, adds detailed position history tracking for both active and closed positions, and provides clear PNL visibility with strategy planning tools.

Risk transparency has been significantly improved with real-time stability pool status and adjustment tracking, a live price impact simulator for large trades, and improved charts showing the SOL to xSOL price relationship.

The most significant structural change in v2 is to sHYUSD itself. Under the new model, sHYUSD no longer takes on xSOL exposure. The previous rebalancing mechanism has been replaced with a market-driven system where the exchange automatically creates MEV opportunities for searchers to buy away fading collateral for USDC, and vice versa, enabling the protocol to re-leverage using its own USDC liquidity when needed. The result is that sHYUSD becomes functionally closer to a standard yield-bearing stablecoin, with some socialized loss from rebalancing activity but no direct SOL volatility exposure. This is a meaningful de-risking of the stablecoin product.

A new ecosystem directory tab consolidates yield strategies across partner protocols like Kamino and Loopscale, showing live APYs and letting users compare options without jumping between apps.

On the asset expansion front, v2 lays the groundwork for non-SOL leveraged assets, with xBTC as the first addition expected in Q2. This diversification reduces Hylo’s single-asset concentration risk and opens the protocol to a broader user base.

Looking Ahead: Q2 2026 Priorities

Q2 priorities are shaped directly by Q1’s lessons. The full v2 launch will roll out the new sHYUSD rebalancing mechanism to all users, removing the SOL volatility exposure that made the stablecoin uncomfortable to hold this quarter and drove a portion of the withdrawals. This single change addresses the core friction that pushed sHYUSD holders out of the protocol: under the new model, stability pool activations no longer convert sHYUSD backing into xSOL, meaning holders retain a stable yield profile regardless of market conditions.

The xBTC launch introduces the first non-SOL leveraged asset, diversifying beyond the single-collateral concentration that amplified Q1’s TVL decline. Asset expansion is not just a product feature but a structural answer to the TVL fragility exposed by a prolonged single-asset drawdown.

The team will also monitor the reconversion of sHYUSD’s remaining ~10% xSOL backing to its target composition, a process that depends on market conditions and the pace at which the protocol can unwind its xSOL position without excessive slippage. Deeper partner integrations through the new Ecosystem Directory round out the quarter, extending Hylo’s reach into the broader Solana DeFi stack through yield strategy aggregation with protocols like Kamino and Loopscale.

Executive Summary

Pear Protocol is a non-custodial trading frontend purpose-built for pair and basket trading in crypto. Rather than operating its own exchange, Pear functions as an aggregation layer that routes orders to underlying perpetual futures venues, primarily Hyperliquid and SYMMIO, borrowing their liquidity and order books while adding a specialized execution layer on top. Users select one or more asset to long and others to short, and Pear handles both legs as a unified position with USDC collateral and up to 60x leverage. The interface surfaces PnL on the spread between the two assets and includes native ratio charting, TWAP execution on the pair ratio, and stop-loss/take-profit logic tied to the spread rather than absolute price. More recently, Pear has introduced Agent Pear, a statistical arbitrage engine that scans all tradable pairs for correlation breakdowns, mean-reversion signals, and hedge ratio shifts, moving the product from a pure execution tool toward an integrated discovery-and-execution platform for relative value strategies.

Q1 2026 was the protocol’s highest revenue quarter at $143K in terminal fees, up 25% quarter-over-quarter, achieved against a backdrop of declining crypto market cap, falling basis rates, and broad deleveraging across perp markets. Active wallets grew 20% to 6,646, and cumulative volume reached $1.47B, suggesting early signs of product-market fit in an environment where relative value strategies are becoming more relevant than pure directional bets. However, core revenue remains well below the estimated $315K quarterly cost base, and the protocol continues to burn cash from its $3.48M treasury, which provides roughly 10-11 quarters of runway. On the product side, the team is developing Pear Pro for institutional traders, a statistical arbitrage vault on Hyperliquid, and exploring a token migration to a more liquid venue, all of which represent potential catalysts for closing the gap to cash flow breakeven.

Income Statement

Q1 2026 marked Pear Protocol’s highest revenue quarter to date, with terminal fee revenue reaching $143K, up ~25% quarter-over-quarter. While revenue has shown volatility over time, the past two quarters have occurred in a relatively weak trading environment, with lower leverage demand and subdued activity across perps markets. In that context, the recent recovery in revenue likely reflects a steady ramp in product usage rather than purely being driven by favorable market conditions.

Pear’s revenue is entirely derived from terminal fees generated through pair trading activity. As such, revenue is still tied to broader market conditions, but the upward trend despite a softer environment suggests early signs of organic growth in usage.

On the cost side, Q1 2026 shows $282K in employee salaries and $33K in software expenses, implying a quarterly cost base of ~$315K. Based on team commentary, prior quarters had a similar expense profile with a modest ramp, indicating that Pear has been operating at a deficit for most of its lifecycle.

A notable line item is Other Income, which has been both large and volatile. This is driven by price fluctuations in HYPE, reflecting treasury mark-to-market changes rather than core operating income. After significant gains in Q2 and Q3 2025, Q4 2025 shows a large loss of $675K, followed by a positive contribution of $370K in Q1 2026, which materially impacts reported profitability but is unlikely to be a stable or repeatable revenue source.

Despite record terminal fee revenue, Pear reported $199K in gross profit in Q1 2026, largely driven by the positive swing in Other Income from the prior quarter’s loss. On a core operating basis (terminal fees vs. expenses), the protocol remains unprofitable, with revenue still below its estimated cost base.

Overall, the income statement suggests Pear is in an early stage where usage is beginning to ramp even in a weaker market environment, but the protocol has not yet reached a level of scale where core revenues can support its fixed operating costs.

Cashflow Statement

Q1 2026 shows -$172K in net cash flow from operating activities, highlighting the gap between reported profitability and actual cash generation. While the protocol reported positive gross profit this quarter, this was largely driven by non-cash Other Income, which does not translate into operating cash flow.

The difference between reported profit and operating cash flow is primarily driven by mark-to-market gains and losses tied to HYPE, which are included in Other Income but excluded from cash flow. As a result, operating cash flow provides a clearer view of the protocol’s underlying sustainability, which remains negative.

On the financing side, the most significant event was the $4.1M raised in Q3 2025, which represents the primary source of treasury inflows. No additional capital was raised in Q1 2026, implying the protocol is currently operating off this existing treasury base. Prior quarters also show some investing activity, including a $500K outflow in Q2 2025. Pear allocates 70% of protocol revenue toward token buybacks and burns, making buybacks a core part of its capital allocation strategy rather than a discretionary expense. In Q1 2026, $100K was spent on buybacks, up from $80K in Q4 2025 and consistent with the growth in terminal fee revenue.

Additionally, Q1 2026 shows -$272K in net treasury deposits, further contributing to the net outflow of capital during the quarter.

Overall, the cash flow profile suggests that Pear is currently burning cash on an operating basis, funded by prior capital raises, while simultaneously returning a majority of revenue to token holders via buybacks. This creates a tradeoff where capital is being distributed before the protocol has reached operating sustainability, making future growth in revenue critical to maintaining both buybacks and runway.

Balance Sheet

Pear Protocol’s treasury stands at $3.48M as of Q1 2026, primarily composed of $1.78M in USDC, $860K in USDH, and $842K in HYPE. The treasury is largely held in stable assets, with ~76% in USDC and USDH, providing a relatively stable base of liquidity, while ~24% is exposed to HYPE price fluctuations.

The current treasury is largely attributable to the $4.1M raise in Q3 2025, which continues to fund operations and capital allocation. Since then, the protocol has been deploying capital through operating expenses and ongoing token buybacks, while also experiencing mark-to-market volatility from its HYPE holdings.

Based on the current cost structure of roughly $315K per quarter, the protocol has approximately 10–11 quarters of runway, excluding any additional volatility in token holdings or changes in spending. This provides a reasonable window for the team to scale revenue.

Overall, Pear’s balance sheet is solid for its stage, with a majority of assets held in stablecoins and sufficient runway to continue development. However, continued operating losses combined with ongoing buybacks mean that extending runway will depend on scaling terminal fee revenue or adjusting capital allocation over time.

KPIs

Q1 2026 saw continued growth across Pear Protocol’s core usage metrics, with trading volume reaching $249.6M, up ~2% quarter-over-quarter and ~38% from Q1 2025. While volume growth has slowed in recent quarters, the protocol has maintained steady expansion despite a relatively weak trading environment.

User growth has been more pronounced. Active wallets increased to 6,646, up ~20% quarter-over-quarter and more than 2.5x since Q1 2025. This suggests that usage is broadening, with more participants engaging with the platform even as overall trading activity stabilizes.

New user wallets totaled 569 in Q1, down from a peak of 848 in Q4 2025. This indicates some moderation in new user inflows, though levels remain above earlier quarters, pointing to continued—albeit uneven—user acquisition.

A notable addition this quarter is $5.2M in open interest, providing a snapshot of active positions on the platform. While historical comparisons are not available, this begins to establish a baseline for measuring capital deployed into Pear’s pair trading strategies.

Overall, the KPI trends suggest that Pear is seeing steady user growth and stable volume, with signs of maturing usage rather than purely hype-driven spikes. The combination of rising active users and sustained volume indicates early product-market fit, though continued growth in both metrics will be key to driving meaningful increases in revenue.

Market Context

These metrics contextualize Pear Protocol’s performance against broader market conditions.

Q1 2026 reflects a continued cooling across key derivatives and macro indicators. Total crypto market cap declined to $2.34T, down from $2.98T in Q4 2025 and significantly off the $4T peak in Q3. At the same time, BTC 3-month annualized basis fell to 2.10%, its lowest level since early 2023, indicating a sharp drop in demand for leveraged directional exposure.

This weakness is also visible in derivatives activity. Hyperliquid volume declined to $660B, down from $770B in Q4 and $1.05T in Q3, while open interest fell to $4.04B from a peak of $9.05B. The contraction in both volume and open interest suggests broad deleveraging and reduced trader participation across perp markets.

Taken together, these indicators point to a market environment characterized by:

• lower leverage demand

• reduced speculative activity

• and weaker directional conviction

This type of environment is typically less favorable for pure directional trading but can create conditions where relative value and pair trading strategies become more relevant, as traders shift away from outright bets and toward positioning on asset dispersion.

Against this backdrop, Pear’s continued growth in users and stable trading volume suggests the protocol is not purely dependent on strong market tailwinds. Instead, it may be beginning to capture usage in a more neutral or uncertain market regime, where its core product—on-chain pair trading—becomes more naturally aligned with trader behavior.

Qualitative analysis

6:08 PMQ1 2026 showed that Pear’s existing user base is relatively sticky. Despite a meaningful contraction in crypto market cap, perp volumes, and leverage demand, trading volume held roughly flat quarter-over-quarter and active wallets continued to grow. This is not a breakout growth quarter, but the fact that usage held up in a weaker environment is a more informative signal than growth driven by market tailwinds.

On the product side, the team is pursuing several initiatives aimed at diversifying revenue and expanding the addressable market. Pear Pro is an institutional-facing version of the terminal designed to bring the pair trading experience into a CEX environment for larger traders. Given that the current terminal already appears to generate sticky usage among on-chain participants, extending that product to a higher-value user segment is a logical next step. The second initiative is a statistical arbitrage vault on Hyperliquid, which would introduce a managed strategy layer and additional trading volume on Pear as well as performance fees. On the token side, the team is exploring migrating liquidity to Aerodrome on Base to deepen trading depth beyond Arbitrum, and is in discussions with legal counsel around evolving the token toward a more explicit ownership model that would give holders a clearer claim on protocol value beyond the current buyback mechanism.

The most notable tension in Pear’s current position is its capital allocation. The protocol allocates 70% of revenue to token buybacks while still operating at a loss, with core revenue covering less than half of its quarterly cost base. At the current burn rate, the ~$100K per quarter directed toward buybacks represents roughly a third of total cash outflows. Halting buybacks until reaching cash flow breakeven would extend runway from approximately 11 quarters to 14, providing meaningful additional time to execute on the product initiatives described above. The team appears focused on iterating toward cash flow positivity, but the current structure prioritizes capital return to token holders before the protocol has reached the scale to sustain it.

Bedrock just went live with its first launchpad partners. A joint venture between Meteora and GVRN AI (a Singapore-based legal technology firm), Bedrock is a permissionless legal framework that incorporates a BVI entity for token projects and places an independent foundation on the cap table with enforceable shareholder rights. It launched with three integrated launchpads: Bags, star.fun, and Collateralize, with more planned.

To understand why Bedrock exists and what problem it is trying to solve, you need to look at what happened in crypto M&A over the past year.

The Problem: 2025’s Acquisition Cycle Exposed Token Holders

Crypto had a record M&A cycle in 2025. Coinbase made nine acquisitions. Pump.fun went on a buying spree. Kraken’s Ink Foundation absorbed an entire protocol. Q3 alone saw 96 transactions totaling over $10 billion, the largest quarter on record. The deals got bigger, faster, and more aggressive.

But through all of it, one group kept getting left behind: token holders.

When a traditional company gets acquired, shareholders receive compensation. The acquirer buys the company, and the shares are the company, so shareholders get bought out. In crypto, that relationship between token and entity has never existed. Tokens sit adjacent to the company. They are not the company. And when the music stops, adjacency pays nothing.

Pantera Capital named this dynamic explicitly in their January 2026 outlook, citing multiple cases where “token-based ecosystems were acquired or restructured without direct compensation to token holders.” They noted this was a key driver behind digital asset equities outperforming tokens in 2025, as investors sought clearer paths to value capture.

Vertex Protocol and Kraken’s Ink Foundation (July 2025)

In July 2025, the Ink Foundation (backed by Kraken) acquired Vertex Protocol’s core technology stack: order book, perpetual futures engine, lending code, and the entire engineering team. The deal was structured to rebuild the DEX exclusively on Ink’s Layer 2 network.

The VRTX token was sunsetted entirely. Holders received an airdrop of just 1% of the future INK token supply, determined by a snapshot taken on July 8. All treasury and non-vested VRTX was burned. At the time of the announcement, VRTX had already fallen roughly 98% from its November 2023 debut, with its market cap collapsing from $77 million to under $4 million. The token dropped an additional 41% on announcement day.

The Ink Foundation absorbed the technology. The Vertex team moved to Ink. Holders were compensated with a sliver of a token that did not yet exist.

Pump.fun and Padre (October 2025)

Pump.fun acquired Padre, a multichain trading terminal with approximately 5% of Solana’s trading bot market share, on October 24. The announcement stated that the PADRE token would “no longer have utility on the platform with no further plans for the future.” The token crashed over 80% within hours.

Initially, no compensation was offered. The community reaction was severe, with widespread accusations of a rug pull. Pump.fun eventually offered a PUMP token airdrop equivalent to the value of PADRE held at a pre-announcement snapshot, claimable by December 30.

As Blockworks noted, this was a goodwill gesture, not a legal obligation. If Pump.fun acquired Padre for $40 million and the PADRE token’s snapshot value was around $5 million, then compensating holders at market was roughly 13% of what they would have received as equity holders in a traditional acquisition. Shareholders would have gotten the full deal value pro rata. Token holders got whatever crumbs the acquirer chose to offer.

Coinbase and Vector.fun (November 2025)

Coinbase’s ninth acquisition of the year was Vector.fun, a Solana-native DEX aggregator built by Tensor Labs. Vector’s technology was folded into Coinbase’s trading infrastructure. The Vector apps were shut down.

The Tensor NFT marketplace and TNSR governance token were carved out and transferred to the independent Tensor Foundation. TNSR holders received no direct compensation from the acquisition. The Foundation burned 21.6% of supply (unvested founder and Labs tokens) and relocked remaining founder allocations for three more years. Marketplace fee revenue was redirected fully to the TNSR treasury.

TNSR saw dramatic price action around the deal. The token surged over 500% in the week prior to announcement, with trading volume spiking from sub-$10 million daily levels to $1.9 billion on November 20, one day before the public announcement. It then dumped 37% on announcement day. The pattern raised significant questions about front-running.

Dragonfly investor Omar Kanji summarized the broader frustration at the time: “TNSR token holders just had their best asset stripped and got ~$0 in return. If this continues, people will just stop buying tokens.”

How Bedrock Works

Bedrock’s answer to this problem is to layer traditional corporate law onto the token launch process. It does not try to make tokens into equity. Instead, it places a legally structured third party on the cap table of the project company, with rights that activate precisely when a founder tries to extract value through equity channels while leaving token holders behind.

The Launch Flow

The process runs in parallel with a normal token launch on any Bedrock-integrated launchpad:

1. The founder creates a token through the launchpad.

2. The founder submits KYC in parallel.

3. The target raise amount is achieved and the token graduates.

4. If KYC clears, a BVI company is incorporated under the Bedrock framework.

5. Upon successful incorporation and documentation, the project is Bedrock-compliant.

The legal infrastructure only activates upon successful incorporation. It is decoupled from the token launch itself. Holding tokens from a Bedrock project confers zero equity rights. The token remains a token. The protection comes from the Foundation’s position as a shareholder in the company.

Equity Allocation

The founder determines how much preference equity Bedrock holds, within set bounds. The minimum is 10% preference shares plus a golden share. The maximum is 30% preference shares plus a golden share. The founder retains at least 70% in ordinary shares and maintains full operational control.

The golden share is the key structural element. It gives the Foundation protective rights that outweigh ordinary shares in specific enforcement scenarios, while leaving the founder free to run the business day to day.

The Takeover Mechanism

This is where Bedrock directly addresses the M&A problem.

The Foundation’s preference equity can be acquired by any party that holds 100% of the project’s total token supply. This is structured as a standing invitation to treat (an expression of willingness to negotiate, not a binding commitment to sell). The counterparty must be an accredited investor and pass KYC/AML/CTF checks.

Since accumulating 100% of a token’s supply on the open market is practically impossible, Bedrock includes a constitutional buyout mechanism modeled on public company mandatory bid codes. Any party that can demonstrate ownership of 30% or more of the total token supply can trigger a mandatory buyout of the remaining tokens.

The buyout price formula is:

(100% - % of tokens surrendered) x Token Total Supply x (7-day TWAP x 1.30)

The 30% premium over the trailing 7-day time-weighted average price is embedded in the company’s constitutional documents. This means if someone wants to acquire the project, they must pay a premium to consolidate both the token supply and the Foundation’s equity position.

For founders, this also creates an exit valve. If the token trades at very low levels, the founder can acquire the supply cheaply and buy back the equity at market prices, effectively unwinding the Bedrock structure.

Enforcement

The Foundation maintains an enforcement fund of up to $2 million. Enforcement activates in cases of founder fraud or bad faith, unauthorized value extraction (acquisition, asset sale, or related-party transaction without engaging the Foundation’s rights), or post-incorporation discovery of criminal conduct.

Enforcement does not activate for project failure, operational disagreements, or token price decline. Bedrock protects against bad actors, not against bad outcomes.

What Founders Get

Beyond the protective structure, Bedrock provides a fully incorporated company with shareholder agreements, proper IP attribution, and ownership set up as part of the launch process. Without it, founders typically spend weeks arranging this themselves, or skip it entirely, creating legal exposure down the line.

How Bedrock Compares to Other Ownership Coin Models

Bedrock is not the only framework on Solana attempting to bridge the token-equity gap. Two other notable approaches are MetaDAO’s futarchy-based ICO model and SOAR’s DRP standard. Each takes a fundamentally different approach.

MetaDAO: Market-Governed Organizations

MetaDAO, backed by Paradigm, implements futarchy: a governance system where prediction markets evaluate proposals rather than token-weighted votes. Anyone can propose an action, and participants trade PASS and FAIL tokens to bet on whether it will increase the organization’s token price. If PASS outperforms by at least 3%, the proposal executes automatically.

MetaDAO expanded from governance into capital formation. Projects launching via MetaDAO use futarchy to control the treasury and mint authority of tokens, meaning the market governs financial decisions rather than a founding team or traditional board. The DAO LLC owns core IP, there is no separate labs entity that can siphon off revenue, and spending requires ongoing market approval.

MetaDAO has facilitated over $33 million in cumulative raises across 11 launches. Protocols like Drift, Jito, and Sanctum have adopted its tools. The model works well for crypto-native organizations that embrace full on-chain governance, but it demands a level of market sophistication and active participation that may not suit every founder or project type.

SOAR: The DRP Standard

SOAR uses its Digital Representation of Participation standard (patent pending). When a company launches on SOAR, only about 5-6% of token supply enters initial circulation. The company signs a senior debt agreement where the debt percentage equals the percentage of tokens in circulation. If 10% of tokens are circulating, the company owes SOAR 10% of its value upon a liquidity event. Minting more tokens increases the debt. Buying back decreases it.

After a 3-month lockup, founders can mint additional tokens but must publicly disclose the amount and purpose 72 hours in advance. There is no governance vote or veto. The check on behavior is transparency and the market’s ability to sell.

As we noted in our January analysis, the model has interesting properties but faces open questions. The contracts between SOAR and launching teams are not currently public, and the platform’s trading volume has settled into five-figure daily levels after an initial burst.

The Key Differences

Where protection comes from: MetaDAO uses prediction markets and DAO ownership of IP. SOAR uses senior debt tied to token circulation. Bedrock uses preference equity and a golden share under corporate law.

Founder freedom: Bedrock gives the most operational freedom, with rights activating only in bad-faith or acquisition scenarios. SOAR creates a soft check via the transparency window. MetaDAO creates the most ongoing friction through continuous market evaluation.

Holder participation required: Bedrock requires none; the Foundation acts on holders’ behalf. SOAR requires monitoring 72-hour minting disclosures. MetaDAO requires active prediction market trading.

M&A specifically: Bedrock is the only framework explicitly designed around acquisition scenarios, with a constitutional buyout mechanism and premium baked into governing documents. MetaDAO makes value extraction difficult by design through DAO IP ownership. SOAR’s debt mechanism creates a financial claim during liquidity events, but enforceability depends on contract terms that are not publicly auditable.

Executive Summary

Strategy’s STRC preferred stock is already one of Bitcoin’s largest marginal buyers, rivaling the entire U.S. spot ETF complex during peak weeks. Unlike ETFs, the flow is one-directional: every dollar of STRC issuance buys BTC, and no amount of STRC selling forces BTC liquidation. But the current scale is just the starting point.

The longer STRC maintains low volatility, the more this engine can grow. Low vol means tighter haircuts in margin frameworks, which means more leverage capacity, which means larger institutional allocations, which means more issuance, which means more BTC buying. Each month of stability extends the track record, opens new buyer pools, and compounds the flywheel. With $21 billion in fresh ATM authorization, the ceiling for this bid is far above where it sits today.

The risk is equally reflexive. STRC can only issue above its $100 par value. A major volatility event that breaks the peg kills the issuance engine, and with it, one of the largest systematic bids in the Bitcoin market. A routine ex-dividend dip in March paused issuance entirely and collapsed weekly BTC purchases from 17,994 to 1,031. A real credit event would do far worse.

STRC volatility is the variable that determines whether this demand source grows or disappears. It belongs on the BTC dashboard.

What STRC Actually Is

STRC, or “Stretch,” is a variable-rate perpetual preferred stock issued by Strategy (formerly MicroStrategy), the largest publicly traded holder of Bitcoin with 762,099 BTC on its balance sheet. The instrument pays a monthly cash dividend at an annualized rate of 11.5% and is designed to trade near its $100 par value. When the price dips below par, the board raises the dividend rate to attract buyers. When it trades above par, the rate can be lowered.

The critical mechanic: Strategy can only issue new STRC shares through its at-the-market (ATM) program when the stock trades at or above $100. Every dollar issued above par becomes capital used to buy Bitcoin. STRC does not exist to pay a dividend. It exists to buy Bitcoin. The dividend is the cost of keeping the machine running.

Since its July 2025 debut, Strategy has raised $5.02 billion through STRC and recently authorized an additional $21 billion in issuance capacity.

STRC Is Rivaling the Entire ETF Complex as a Marginal BTC Buyer

During the week of March 9-15, 2026, STRC ATM sales generated $1.18 billion in proceeds. Strategy used those proceeds (combined with MSTR common stock sales) to purchase 17,994 BTC at an average price of $70,946. Over that same period, the combined net inflows across all 12 U.S. spot Bitcoin ETFs totaled approximately $763 million.

STRC alone was a larger marginal buyer of Bitcoin than BlackRock’s IBIT, Fidelity’s FBTC, and every other spot ETF combined.

The comparison undersells the point, because ETF flows are bidirectional. On January 29, the entire ETF complex posted negative $817.8 million. When ETF investors sell, authorized participants redeem shares and sell Bitcoin into the market. STRC does not have this mechanic. When STRC holders sell their shares, they sell into the equity market. Strategy does not sell Bitcoin to meet redemptions because there are no redemptions. The BTC goes in and does not come out.

This is a structurally one-directional flow. Every dollar of STRC issuance creates a Bitcoin bid. No amount of STRC selling creates a Bitcoin ask.

Volatility Is the Circuit Breaker

STRC’s rolling volatility is the single best real-time indicator of whether this bid is active. The connection runs deeper than ATM mechanics.

In collateral and leverage markets, the lower an asset’s observed volatility, the more leverage it can support. Prime brokers, clearinghouses, and repo desks set haircuts based on historical volatility using VaR models calibrated to rolling lookback windows. An asset with 2% rolling vol gets a smaller haircut than one with 18% vol. Smaller haircut means more borrowing capacity per dollar of collateral. More borrowing capacity means more capital deployed against the position.

As STRC’s 30-day rolling vol has compressed from 18% to 2%, every institution holding it gets progressively more room to size up. Funds running leveraged carry trades can take larger positions. Treasuries can justify larger allocations because risk models show less potential drawdown.

But the more important mechanism is temporal. Each additional month that STRC holds near par at low vol extends the track record that margin models use. Compliance teams get more comfortable. New buyer pools open up that were waiting for exactly this kind of demonstrated stability before entering. The instrument graduates from “novel crypto-adjacent product” to “yield instrument with a proven volatility profile,” and each graduation unlocks a new tier of capital.

Each new tier of capital creates more ATM issuance, more BTC buying, a stronger balance sheet, better credit quality, and more STRC stability. The reflexive loop feeds on its own history. The longer it runs, the stronger it gets.

This is why STRC vol is the master variable. It governs ATM issuance pace directly, leverage and sizing capacity indirectly, and the rate of new capital formation over time.

What to Watch

STRC vol belongs on the BTC dashboard alongside ETF flows and exchange balances.

30D rolling volatility is the primary signal. Any sustained move above 5% would indicate the rate adjustment mechanism is losing its grip and issuance capacity is being impaired.

The $99-$101 trading band is the early warning. BitcoinQuant tracks the percentage of shares trading within this range, which has been near 100% recently. A breakdown in this band precedes an issuance pause.

Daily volume vs. 30D average signals demand intensity. The current 30D average is $221 million. Sessions well above average suggest accelerating institutional demand. Sessions well below suggest the bid is thinning.

Ex-dividend recovery time is the recurring stress test. Each monthly ex-dividend date pushes STRC below par temporarily. The speed of recovery (currently 9-10 trading days) measures underlying demand strength. If recovery starts taking 15 or 20 days, the issuance window shrinks and the BTC bid weakens proportionally.

Executive Summary

@Polymarket is rolling out taker fees across nearly all market categories on March 30, 2026, expanding a fee structure that currently only applies to Crypto and Sports markets. Using on-chain data from Dune Analytics and Polymarket’s published fee parameters, we estimate that the platform will generate approximately $1.2M in daily gross fee revenue at current volume levels (~$160M/day in taker volume). After accounting for maker rebates and referral payouts, protocol net revenue is estimated at $573K per day, or roughly $209M annualized. Early evidence from the Crypto and Sports fee rollouts suggests the impact on trading volume is modest, with Sports markets showing no volume decline and Crypto markets seeing only a slight pullback that falls within normal variance.

How Polymarket’s Fee Structure Works

Polymarket uses a dynamic taker-only fee model. Makers (liquidity providers who place resting orders) pay nothing and instead receive rebates. Takers (traders who hit existing orders) pay a fee that varies based on the probability of the market they are trading. The fee is calculated using the formula: fee = C x p x feeRate x (p x (1 - p))^exponent, where C is the number of shares and p is the share price between 0 and 1.

This creates a bell-shaped fee curve that peaks at 50% probability, where uncertainty is highest, and drops toward zero at the extremes. A trade at 95 cents (95% implied probability) incurs a negligible fee, while a trade at 50 cents pays the maximum rate. Geopolitics and world events markets remain completely fee-free.

The peak effective fee rates by category after the March 30 rollout are as follows:

• Crypto — 1.80%

• Sports — 0.75%

• Finance — 1.00%

• Politics — 1.00%

• Economics — 1.50%

• Culture — 1.25%

• Weather — 1.25%

• Tech — 1.00%

• Other/Mentions — 1.25–1.56%

• Geopolitics — 0%

Fee Distribution

Collected taker fees are split three ways. First, a category-specific portion (20% for Crypto, 25% for most categories, 50% for Finance) is allocated to the Maker Rebates Program, paid daily to liquidity providers in proportion to their filled maker volume. Second, Polymarket runs a referral program that pays referrers 30% of gross taker fees generated by direct referrals and 10% from indirect referrals. The remaining balance is retained by the protocol. In practice, since not every trade is referred, the protocol’s actual share is likely higher than the base 47% implied by the math.

Early Evidence: Volume Impact of Fees

Fees first went live on Crypto markets on January 5, 2026, initially limited to 15-minute contracts and later expanded to all crypto timeframes by March 6. Sports market fees launched on February 18. This gives us roughly two months of data on Crypto and one month on Sports to assess the volume impact.

Crypto markets saw a brief volume spike around the fee expansion in early March, reaching $100-120M in daily volume, before settling back to a range of $73-100M. While this represents a modest pullback, it falls within the normal variance of Crypto market volumes and is difficult to attribute definitively to fees after only 18 days of data. Sports markets, on the other hand, have been completely unfazed. Volume has actually increased from the $100-150M range to $150-250M per day since the fee introduction.

One likely explanation for the divergence is the fee differential: Crypto’s peak rate (1.56%, rising to 1.80%) is roughly 3.5x higher than Sports (0.44%, rising to 0.75%). The remaining categories being added on March 30 have peak rates that fall between these two, suggesting any volume impact is likely to be minor. Additionally, the maker rebate program creates a structural incentive to backfill any liquidity gaps, since market makers are now being paid to provide liquidity rather than doing so for free.

Revenue Estimates

Methodology

To estimate fee revenue, we first needed to understand where on the probability curve trades are actually occurring. We queried the last 7 days of on-chain trade data from Polymarket’s CTF Exchange contract on Polygon using Dune Analytics, calculating the volume-weighted average position on the fee curve relative to the theoretical maximum at 50% probability. Because the fee formula uses different exponents by category, which changes the shape of the curve, we calculated this separately for each exponent value. For exponent 1 categories (Crypto, Sports, Politics, Finance, Culture, Tech), the average trade captures 70.1% of the peak fee. For exponent 2 categories (Other/General, Mentions), it drops to 59.7% since the steeper curve penalizes off-center trades less. For exponent 0.5 categories (Economics, Weather), it rises to 80.2% since the flatter curve means traders pay closer to the max rate even at skewed probabilities.

We then applied the appropriate factor to each category’s peak fee rate and multiplied by recent daily taker volume. Taker volume (~$160M/day) represents the actual USDC changing hands per trade, roughly half of the ~$350M notional volume figure that Polymarket reports, which counts the full $1 face value of each share traded.

Results

Across all categories, the blended effective fee rate comes to approximately 0.76% of taker volume. At current levels, this produces the following estimated daily and annualized figures:

One important caveat: the referral figure represents the maximum payout assuming every trade comes through a referral link. In practice, the protocol’s net take is likely meaningfully higher than $209M annualized, with the true figure falling somewhere in the $209-342M range depending on referral program adoption.

Competitive Context

The implementation of fees on existing markets has had a slightly negative to negligible effect on the growth trajectory, and the maker rebate structure creates built-in incentives to compensate for any minor shortfall in liquidity. For context, @Pumpfun currently generates roughly $1M per day in net protocol revenue and @HyperliquidX sits around $2M per day. Post-transition, Polymarket’s estimated net protocol revenue of ~$580K per day (after maker rebates and referral payouts) puts it at roughly 60% of pumpfun and 30% of Hyperliquid on a like-for-like basis. Notably, the referral payout assumption is a ceiling since not every trade is referred, so the real protocol take likely sits closer to $580K-$950K per day depending on referral adoption. At the upper end of that range, Polymarket is already approaching pumpfun’s net revenue. If Polymarket continues even half of its recent growth trajectory, which has seen roughly 200% volume growth over the past six months, it could match pumpfun on a net basis and close the gap with Hyperliquid by September.

Among large cash-producing applications in crypto, Polymarket’s growth profile currently looks among the strongest. The combination of expanding product categories, mainstream media integrations (including the X/Twitter partnership), regulatory normalization through the ICE investment, and a fee structure designed to minimize friction while maximizing maker incentives positions the platform well for sustained revenue generation.

Sources

• https://dune.com/filarm/polymarket-activity

• https://dune.com/queries/6903092/?utm_source=share&utm_medium=copy&utm_campaign=query

• https://blockworks.com/analytics/polymarket

• https://docs.polymarket.com/trading/fees#sports

• https://defirate.com/prediction-markets/

Introduction

Bittensor’s native token TAO trades at approximately $275 with a $2.6 billion market cap and $5.8 billion FDV. The project has institutional backing from Grayscale (S-1 filed December 2025 for a NYSE-listed ETF), a public endorsement from NVIDIA CEO Jensen Huang, and a compelling supply narrative: a 21 million hard cap with a Bitcoin-style halving schedule. The first halving in December 2025 cut daily emissions from 7,200 to 3,600 TAO. Subnet count has grown from 32 to 128 in a year, and Templar’s Covenant-72B training run proved decentralized compute can produce benchmark-competitive language models.

This report does not dispute any of that. It examines whether the network’s economics can generate demand-side revenue at a scale that justifies current valuations, and how its competitive positioning holds up against centralized providers and self-hosted compute.

How Value Flows Through the Network

Bittensor has four participant classes. Subnet owners create specialized AI marketplaces and receive 18% of their subnet’s TAO emissions. Miners perform AI work (inference, training, data processing) and receive 41%, totaling approximately 1,476 TAO per day (~$406,000 or $148 million annualized). Validators score miner output and receive 41%. Stakers allocate TAO into subnet liquidity pools, receiving subnet-specific alpha tokens. Under the Taoflow model, a subnet’s emission share is determined by its net TAO staking inflows. Subnets with negative flows receive zero emissions.

TAO is the universal entry currency. Mining registration, validator staking, subnet token purchases, and service payments all require it. Subnet activity theoretically creates structural demand for the base token.

The Demand Side

Supply Transparency vs. Demand Opacity

Bittensor’s supply economics are transparent. Daily emissions of 3,600 TAO are split programmatically: 41% to miners, 41% to validators, 18% to subnet owners. The halving schedule is hard-coded. Staking ratios (~70% of supply), emission shares, and flow data are all on-chain. The top 10 subnets control roughly 56% of total emissions.

The demand side has no equivalent transparency. There is no aggregated dashboard tracking external revenue by subnet. AI service delivery (inference requests, compute jobs, training calls) happens off-chain and is not recorded on the blockchain. Investors infer demand from proxy metrics: staking flows, subnet token prices, and self-reported figures from individual teams. This opacity is structural, not temporary. The blockchain records token movements, not API calls.

What follows is the most complete demand-side picture available as of March 2026.

Chutes (SN64): The Subsidy Behind the Savings

Chutes commands 14.4% of network emissions, the highest of any subnet. Developed by Rayon Labs, it offers serverless inference on open-source models (DeepSeek, Mistral, LLaMA) at prices reported to be 85% below AWS and 10-50% below Together AI. Usage metrics are the ecosystem’s strongest: 400,000+ users (100,000+ via API), 5M+ daily requests, 9.1 trillion tokens processed, and daily token generation that surged from 6.6 billion to 101 billion over three months. Chutes ranks as a top inference provider on OpenRouter, outperforming some centralized competitors on specific models.

The pricing, however, does not reflect operational efficiency. It reflects subsidy. At 14.4% of emissions, Chutes receives approximately 518 TAO per day, worth roughly $142,000 or ~$52 million annualized. This flows to the subnet's miners, validators, and owner. Against that, estimated external revenue is approximately $1.3-2.4 million per year (the higher figure self-reported by the Chutes team to DL News in October 2025; the lower from a March 2026 estimate, neither independently audited). The protocol subsidizes the subnet's operations at a ratio of roughly 22-40:1. For every dollar customers pay, the network contributes $22-40 in TAO emissions.

The implications become clear when you model what pricing would look like without the subsidy. Chutes processes approximately 101 billion tokens per day. If miners had to cover their costs from customer revenue alone, they would need to recoup that ~$142,000 daily from token throughput. That implies an unsubsidized price of roughly $1.41 per million tokens.

Current market rates for the same open-source models: Together.ai prices LLaMA 3.3 70B Turbo at approximately $0.88 per million tokens. DeepSeek V3 runs at $0.40-0.80 across competitive providers. Smaller models go as low as $0.18 per million. At $1.41 per million, unsubsidized Chutes pricing would be 1.6-3.5x more expensive than centralized alternatives. The cost advantage does not narrow. It inverts entirely. The 85% savings that headline every Chutes discussion are funded by TAO holders through inflation, not by any structural efficiency of decentralization.

When emissions halve again (projected late 2026 or 2027), either pricing roughly doubles, miners leave the subnet, or the gap between subsidy and revenue widens further. The obvious rebuttal is that this resembles standard startup bootstrapping: subsidize early to build network density, then raise prices once users are locked in. Uber, DoorDash, and AWS all ran this playbook. The difference is that those companies built switching costs during the subsidy period (proprietary platforms, driver networks, enterprise integrations). Bittensor subnets build none. The models are open source, the APIs are standard, and users can migrate to any provider serving the same weights with zero friction. When the subsidy shrinks, there is no lock-in to prevent churn.

Rayon Labs also operates SN56 (Gradients) and SN19 (Nineteen), together commanding ~23.7% of total emissions. Neither has disclosed external revenue figures. A single team controls nearly a quarter of the network’s incentive distribution.

Targon, Templar, and the Rest

Targon (SN4) is the highest-revenue subnet. Operated by Manifold Labs ($10.5M Series A), it provides confidential GPU compute to enterprise customers. Estimated annualized revenue is ~$10.4 million against a ~$48 million valuation, implying a 4.6x revenue multiple. This is the most grounded valuation in the ecosystem. However, the $10.4 million is a projection cited across multiple analyst reports, not an audited number. Multiple researchers have flagged the absence of a live revenue dashboard excluding emissions as a transparency concern.

Templar (SN3) completed Covenant-72B (72B parameters, 1.1T tokens, 67.1 MMLU) and carries a $98 million market cap. It generates zero external revenue. Training APIs and enterprise sales are described as “in motion” but no paid product has launched.

The remaining 120+ subnets either have no reported external revenue, are pre-product, or operate primarily on emission capture. Celium (SN59) claims revenue but has published no figures. Nineteen AI (SN19) has disclosed no data.

The Aggregate Picture

Total identifiable demand-side revenue across the entire network is approximately $3-15 million annually, with the range reflecting the difference between verified revenue (Chutes’ ~$2.4M) and unaudited projections (Targon’s ~$10.4M). A single subnet’s emission subsidy (~$52M for Chutes alone) exceeds the upper bound of what the entire network generates in external revenue.

Against a $2.6 billion market cap, this implies a ~175-200x revenue multiple. Against FDV of $5.8 billion, ~400x. Centralized AI infrastructure companies (CoreWeave, Lambda) were valued at 15-25x forward revenue in recent rounds. High-growth SaaS rarely sustains above 50x. Bittensor’s implied multiple is 4-10x higher than the most aggressively valued comparable in either crypto or traditional infrastructure.

The gap between valuation and demand-side economics reflects a market pricing TAO almost entirely on supply-side scarcity (halving, staking lock-up), institutional catalysts (Grayscale ETF, exchange listings), and AI sector sentiment rather than demonstrated economic productivity. These are real price drivers. But they are distinct from a thesis that Bittensor is generating sustainable value as an AI services network.

The Pricing Vise

Subnets face margin compression from two directions simultaneously.

Self-hosting caps pricing from above. Every model on Bittensor is open source. Weights are on Hugging Face. A single H100 serves a 70B model at ~$40-50/day all-in. Tools like vLLM and Ollama have made local deployment trivial. NVIDIA’s roadmap (Blackwell, Rubin) is designed to reduce inference cost per token by orders of magnitude. Self-hosting offers unlimited inference with no token friction, no counterparty risk, and no network dependency. Any organization with sufficient volume is already cheaper running locally.

Hyperscalers compress from below. Microsoft, Google, Amazon, and Meta collectively invested over $200 billion in AI capex in 2025. They have first-priority hardware allocation, purpose-built data centers, existing enterprise relationships, and the ability to subsidize AI from adjacent cash flows. Bittensor’s entire annual incentive budget (~$360 million) is less than what Microsoft spends on AI infrastructure in a week. Specialized providers (Together.ai, Fireworks, Groq) compete on the same open-source models with VC-subsidized pricing.

Subnets must price between these two boundaries while absorbing costs unique to decentralization: token friction, validator overhead, subnet owner fees, and network latency.

The Moat Problem

If a subnet develops a valuable service, the underlying model and methodology are publicly available by design. Covenant-72B is Apache-licensed. SparseLoCo is on arXiv. Any competitor can replicate the approach without participating in the TAO economy.

Traditional moats (proprietary technology, network effects, switching costs, brand) do not apply. The technology is open. Network effects accrue to TAO, not individual subnets. Switching costs are zero when every provider serves the same weights. The community argues the incentive mechanism is the moat, but that requires the emission budget to remain large enough to attract compute, and it shrinks with every halving.

What TAO Is Pricing

At $2.6 billion, TAO is not priced on demand-side fundamentals. $3-15 million in annual revenue does not support it under any conventional framework. The market is pricing Bitcoin-like scarcity, the Grayscale ETF catalyst, AI sector rotation, and long-term optionality on decentralized AI. These are legitimate speculative factors. They are also entirely supply-side and sentiment-driven.

A TAO position based on scarcity and narrative may perform well regardless of demand economics. A position based on Bittensor becoming a meaningful AI services network requires evidence that does not yet exist and faces structural headwinds that may prevent it from materializing. Investors should be clear about which thesis they hold.