Crypto protocols spent more than $1.4 billion on token buybacks in 2025, according to [CWallet research](https://cwallet.com/blog/2025-token-buybacks-near-1-4-billion-dominates-top-10-projects/). The aggregate result: most tokens subject to buyback programs remain flat or sharply lower. The data ...
"In markets with heavy emissions, short-term buybacks do not reset how sellers price risk. Tokens unlocked today are sold at today's price, not at some future value implied by ongoing repurchases." — Anatoly Yakovenko, Co-founder, Solana
Crypto protocols spent more than $1.4 billion on token buybacks in 2025, according to CWallet research. The aggregate result: most tokens subject to buyback programs remain flat or sharply lower. The data challenges the premise that buybacks alone constitute a credible value-return mechanism for token holders.
Yet the practice is accelerating, not retreating. In Q1 2026, Optimism, Pyth Network, Uniswap, Pendle, and Treehouse each launched or expanded buyback programs. Aave paused its $50M annual program following the rsETH incident in April 2026. Jupiter halted its program entirely after $70 million in repurchases failed to offset a 150% increase in circulating supply. The divergence in outcomes — Hyperliquid's 97% fee-to-buyback model versus Jupiter's $70M failure — reveals that mechanism design and emission schedules matter far more than headline buyback volumes.
This report maps the current state of protocol buyback programs, identifies which structures deliver value to token holders versus corporate entities, and assesses whether the buyback trend represents genuine capital return or cosmetic treasury management.
Development activity around token buyback infrastructure is increasing, though concentrated in early-stage tooling rather than production-grade protocol code.
Autonomous Buyback Libraries. The most notable repo is loothero/autonomous_buyback, a Cairo library for autonomous token buybacks via Ekubo TWAMM on Starknet. The repo received its most recent update on January 23, 2026, with a v2 buyback component featuring enhanced ERC20 burn support and stream-based token distribution. The codebase includes 18 comprehensive burn tests covering approval mechanics, allowance deduction, and infinite approval patterns. This represents on-chain infrastructure for protocols to automate buybacks without manual intervention — a direct response to criticism that many buyback programs are executed inconsistently.
Treasury Management Tooling. Alex000115/treasury-tax-harvester provides a Uniswap V3 integration module for converting protocol fees into native tokens. Updated March 2026, the repo is a professional-grade treasury management module designed for automated buyback execution. Separately, the Multi-Sig-Treasury-Vault pattern — a multi-signature wallet interface for DAO treasury management — has seen five forks in April-May 2026, indicating growing demand for secure treasury infrastructure.
AI x Buyback Convergence. BagsAI Agent Forge on Solana demonstrates the intersection of AI agents and buyback mechanics — each deployed AI agent receives its own token with automatic fee splitting, token buybacks, and holder rewards baked into the contract layer. Updated April 2026, it reflects a growing pattern of embedding buyback mechanics directly into AI agent token launches.
The GitHub signal is clear: buyback infrastructure is moving from ad-hoc OTC desk arrangements toward programmable, on-chain, autonomous execution. The tooling remains nascent, but the trajectory points toward buybacks becoming a standard protocol primitive rather than a discretionary treasury decision.
According to The Block and Bitget research, the top 10 crypto buyback programs in 2025 accounted for 92% of the $1.4 billion total. The concentration is stark:
| Protocol | 2025 Buyback Volume | Token Performance | |----------|-------------------|-------------------| | Hyperliquid | $645M | Outperformed sector | | LayerZero | $150M | Mixed | | Pump.fun | $138M | N/A (no token) | | Jupiter | ~$70M | -89% from peak | | Aave | ~$50M | Moderate |
Per CryptoSlate, excluding Hyperliquid, the average return for tokens with buyback-and-burn programs was -56%. The one number that determines effectiveness: the ratio of buyback volume to new token emissions. When emissions outpace buybacks — as in Jupiter's case — the program functions as a subsidy to sellers, not a return to holders.
The WisdomTree analysis notes that DeFi protocols are explicitly adapting TradFi's buyback playbook, but without the regulatory framework (SEC Rule 10b-18) or the accounting discipline (EPS accretion targets) that gives corporate buybacks structural purpose. The result is a fragmented landscape where buyback announcements function primarily as signaling mechanisms.
Hyperliquid channels 97% of trading fees into its Assistance Fund, which automatically purchases HYPE from the open market, per Tokenomics.com. The protocol generated $65 million in monthly holder revenue as of early 2026, with a record $6.84 million single-day revenue on February 5, 2026, according to AMBCrypto. On that day alone, 160,750 HYPE were repurchased.
In December 2025, the Hyper Foundation proposed burning 37 million HYPE (~$1 billion, 13% of circulating supply) through a validator vote establishing "social consensus" that Assistance Fund tokens are permanently removed, per The Defiant. More recently, Hyperliquid posted $5.23 million in daily revenue on April 17, 2026, according to CryptoTimes.
The model works because Hyperliquid generates organic revenue at scale relative to its float. Over 40.5 million HYPE have been accumulated since the program began in late 2024.
Optimism governance approved redirecting 50% of Superchain sequencer revenue to OP buybacks in January 2026, with 84.4% of votes in favor, per The Block. Superchain sequencers generated approximately 5,900 ETH ($8 million annualized at then-current prices) over the prior year, per CoinDesk.
The Foundation partners with an OTC provider for monthly ETH-to-OP conversions, with a floor mechanism: the program pauses if monthly revenue drops below $200,000. The 12-month pilot began in February 2026 per Optimism's blog.
Aave's Aavenomics buyback program, launched in 2025, acquired over 205,000 AAVE (1.28% of total supply) through weekly purchases of $250K-$1.75M, per the Aave governance forum. The DAO approved a permanent $50M annual budget, per DL News. However, following the rsETH bridge incident on April 18, 2026, buybacks were paused on April 19. A proposal to reduce the annual budget from $50M to $30M was already under consideration prior to the pause.
Jupiter spent over $70 million on JUP buybacks using roughly half of protocol fee revenue, per crypto.news. By January 2026, JUP traded near $0.20-$0.22, down approximately 89% from its peak. The core problem: circulating supply increased by ~150% since launch, with 53 million JUP unlocking monthly through June 2026. Jupiter co-founder Ong acknowledged the futility and halted the program, pivoting to growth incentives, per Bitget. The planned 2026 airdrop was cut from 700 million to 200 million JUP.
The UNIfication proposal passed in December 2025 with 125 million votes in favor and only 742 dissenting, per The Defiant. Uniswap activated protocol fees (0.05% on v2 pools) and a programmatic UNI burn mechanism. A 100 million UNI burn (~$596M notional) was executed in early January 2026, representing retroactive value that would have accrued to holders had the fee switch been active since launch, per KuCoin research.
Since rollout, Uniswap has burned more than $5.5 million worth of UNI, implying a $34 million annualized pace. In February 2026, UNI jumped 15% on a governance vote to expand the fee switch to eight L2 networks, which could add $27 million in annualized revenue, per CoinDesk and Coin Metrics.
Pendle executed one of the more structurally significant governance transitions in Q1 2026. The protocol retired its vePENDLE model — which required multi-year token lockups — and replaced it with sPENDLE, a liquid staking token with a 14-day unstaking period and an instant-exit fee option, per BanklessTimes and The Block.
Under the new tokenomics, up to 80% of protocol revenue is directed to PENDLE buybacks distributed to sPENDLE holders. A snapshot on January 29 captured existing vePENDLE positions, granting holders a loyalty boost of up to 4x, decaying linearly over two years.
The challenge: Pendle's revenue collapsed 87.6% from $4.44M in August 2025 to $552K in March 2026 due to DeFi yield compression, per AInvest. At $34M annualized revenue (as of April 2026), the buyback allocation amounts to roughly $21 million annually. The protocol claims a 4.4x deflationary surplus over emissions, but the revenue trajectory raises sustainability questions.
Treehouse, with $294 million in deposits, approved allocating 50% of fees from its tETH yield product to recurring TREE token purchases, per DL News. The governance proposal passed with 99.59% community support. TREE doubled in price following the announcement but remains 35% below its August all-time high of $0.40.
The Treehouse case illustrates a pattern: smaller protocols use buyback announcements as catalysts for price recovery, with the DAO treasury retaining purchased tokens for future governance-directed deployment rather than burning them. The value accrual is indirect — tokens sit in the treasury rather than being removed from supply.
Maple Finance replaced inflationary SYRUP staking rewards with a revenue-funded buyback mechanism following the MIP-019 governance vote in October 2025, per Yahoo Finance. Twenty-five percent of protocol revenue now flows to the Syrup Strategic Fund (SSF), which purchases SYRUP on the open market. SYRUP jumped 16% after Maple revealed a $2 million token buyback, per The Defiant.
Maple's active loans grew 8.4% to $2.4 billion, with 70% from syrupUSDC. The protocol targets $100M ARR by end of 2026, per VaasBlock. Maple's partnership with Aave to integrate syrupUSDC and syrupUSDT into Aave's lending markets creates a cross-protocol revenue loop — Maple earns fees from institutional lending, buys back SYRUP, while Aave earns from syrupUSDC deposits.
Pyth launched its PYTH Reserve program, deploying 33% of DAO treasury balance monthly to purchase PYTH on the open market, per The Block. Initial monthly buybacks are $100K-$200K from a ~$500K treasury. The program is small in absolute terms but structurally significant: Pyth is among the first oracle networks to implement direct revenue-to-buyback mechanics. The Pyth DAO has published monthly purchase reports for January, March, and April 2026, providing unusual transparency. Pyth Pro surpassed $1M ARR in its first month, per FinanceFeeds.
The critical question for token holders: who benefits from protocol revenue — token holders, or the corporate entities behind the protocols?
Foundation-Mediated Buybacks. Optimism's buyback is executed by the Optimism Foundation through an OTC provider. The Foundation retains discretion over execution windows, provider selection, and program continuation. Jito's buybacks are managed by its Community Stewardship Directorate. In both cases, foundations — not token holders — control the capital allocation.
DAO-Directed Programs. Aave and Uniswap route buyback decisions through DAO governance, providing more direct token-holder control. However, the Aave governance forum reveals that even DAO-directed programs can be paused unilaterally when risk events occur (the rsETH incident).
Protocol-Embedded Mechanics. Hyperliquid's Assistance Fund operates autonomously — 97% of fees flow to buybacks without governance votes or foundation discretion. This is the closest approximation to a dividend-equivalent mechanism in crypto. Pendle's sPENDLE model similarly embeds buyback allocation at the protocol level.
The Equity Layer. Notably absent from most buyback discussions: the equity layer. Uniswap Labs (the company) generates revenue from its front-end fee (0.15% on swaps), entirely separate from the protocol fee that feeds UNI burns. Aave's service providers receive separate DAO grants. The foundation or labs entity typically holds significant token allocations subject to vesting. In practice, protocol revenue is split between token buybacks (visible) and corporate operational costs (less visible), with token holders bearing dilution from ongoing emissions while corporate entities receive guaranteed operational funding.
The data reveals a clear hierarchy of value-return effectiveness:
Tier 1 — Direct, Autonomous, Revenue-Backed: Hyperliquid (97% fee-to-buyback, autonomous execution, $65M monthly). Token holders receive near-direct revenue exposure.
Tier 2 — Protocol-Embedded with Revenue Link: Pendle sPENDLE (80% revenue to buybacks), Uniswap (fee switch + burn). Value accrues to token holders through supply reduction, but revenue volatility introduces uncertainty.
Tier 3 — Foundation/DAO-Mediated: Optimism (50% sequencer revenue, OTC execution), Aave (weekly purchases, subject to pause), Maple (25% revenue to SSF). Value return is real but intermediated by corporate entities with discretion to pause or modify.
Tier 4 — Structurally Challenged: Jupiter ($70M spent, -89% token performance, halted). Buyback volume insufficient relative to emission schedule.
The pattern: buybacks work when revenue scale exceeds emission dilution, execution is automated rather than discretionary, and no corporate intermediary can unilaterally pause the program.
The crypto buyback trend of 2025-2026 represents a genuine structural shift toward connecting protocol revenue with token holder value — but the mechanism is only as good as its implementation. The $1.4 billion spent in 2025 produced a -56% average return (ex-Hyperliquid), demonstrating that buyback announcements are not a substitute for sound tokenomics.
The protocols that deliver real value to token holders share three characteristics: revenue that materially exceeds emissions (Hyperliquid), autonomous on-chain execution that cannot be paused by corporate entities (Hyperliquid's Assistance Fund, Pendle's sPENDLE), and transparent reporting of buyback activity (Pyth's monthly reports). Protocols lacking any of these — particularly those where foundations retain discretion over execution — are running what amounts to corporate treasury management programs marketed as token holder value return.
The next twelve months will likely separate sustainable buyback models from cosmetic ones. As the infrastructure matures — Cairo libraries for autonomous buybacks, Uniswap V3 integration modules — the barrier to implementing credible, on-chain, non-discretionary buyback programs will drop. Protocols that fail to adopt autonomous execution will face increasing scrutiny from token holders who have seen $1.4 billion spent with little to show for it.