Crypto protocols spent over $1.4 billion on token buybacks in 2025, per a [DWF Labs](https://www.dwf-labs.com/research/547-token-buybacks-in-web3) research report, with the top 10 programs capturing 92% of total expenditure. The trend has accelerated into 2026: Hyperliquid now runs $1 million per...
"The market doesn't seem to care about projects buying their tokens back off the market." — Amir Haleem, Former CEO, Helium
Crypto protocols spent over $1.4 billion on token buybacks in 2025, per a DWF Labs research report, with the top 10 programs capturing 92% of total expenditure. The trend has accelerated into 2026: Hyperliquid now runs $1 million per day in automated buybacks, Aave has locked in a permanent $50 million annual program, and Optimism governance approved routing 50% of Superchain revenue to OP purchases. Uniswap's fee switch, activated in December 2025, has burned over $5.5 million worth of UNI through Q1 2026.
Yet the data presents a paradox. Jupiter spent $70 million buying back JUP while the token fell 89%. Helium paused its program entirely after concluding it produced no measurable price impact. These divergent outcomes expose a structural question: under what conditions do buybacks create genuine value for token holders versus simply burning cash?
This report maps the emerging buyback landscape across DeFi, L1/L2 infrastructure, and niche protocols, analyzing where revenue flows, who controls the treasury, and whether token holders are positioned to capture value — or merely watch it evaporate.
Development activity around buyback infrastructure has moved from theoretical to production-grade. A search of GitHub repositories updated in 2026 reveals several relevant projects:
treasury-tax-harvester (Alex000115) — A Solidity-based module integrating with Uniswap V3 for automated protocol fee conversion into native tokens. The repo's three initial commits in March 2026 implemented a TaxHarvester.sol contract with swap parameter calculation tooling. Zero stars, zero forks — indicating this is early-stage infrastructure, but the architecture mirrors what production protocols are deploying internally.
autonomous_buyback (loothero) — A Cairo library for autonomous token buybacks via Ekubo TWAMM (Time-Weighted Average Market Making) on Starknet. The v2 component, committed January 2026, added ERC20 burn support and distribution order management with 18 comprehensive tests. This signals that buyback automation is now reaching L2 ecosystems beyond EVM.
M0 Foundation's TTG (Two Token Governance) — 11 stars, last pushed May 30, 2026. This governance framework separates voting power from economic claims using two distinct tokens. The architecture represents a design pattern where governance rights and value accrual are structurally decoupled — relevant as protocols debate whether buyback recipients should also control treasury decisions.
LayerZero Analytics Dashboard (Twojekrypto) — Updated June 7, 2026 (today), this repo tracks multi-chain holder flows, tokenomics, vesting, and buybacks for the ZRO token. The real-time update cadence suggests active community demand for buyback transparency tooling.
The GitHub data points to a maturing infrastructure layer: protocols are no longer debating whether to implement buybacks but building the tooling to execute them autonomously and transparently.
The DeFi buyback wave of 2025-2026 can be segmented into three distinct tiers based on execution scale and structural design:
Tier 1: Automated, fee-funded, continuous — Hyperliquid, Aave, Uniswap
Tier 2: Governance-approved, periodic, treasury-funded — Optimism, Lido, CoW DAO
Tier 3: Experimental, niche, or paused — Jupiter, Helium, Treehouse, Jito
According to DWF Labs, Hyperliquid alone accounted for roughly $645 million of the $1.4 billion in buybacks executed in 2025 — nearly half the market total. The concentration is striking: true buyback volume is dominated by protocols with sustainable fee revenue, not by the number of programs announced.
Hyperliquid's Assistance Fund routes 97% of trading fees into automated open-market HYPE purchases, per CryptoTimes reporting. By May 2026, cumulative buybacks exceeded $1.3 billion, with the Fund holding approximately 28.5 million tokens. At $800 million in annualized fees and $5.8 billion TVL as of June 3, 2026, the buyback rate runs approximately $1 million per day — an annualized yield of roughly 7% of market cap.
The corporate structure matters here. Hyperliquid has no venture investors, no foundation token allocation, and no equity overhang. All protocol revenue flows to the buyback mechanism. Bitwise launched a spot HYPE ETF (NYSE: BHYP) on May 15, 2026, which accumulated approximately $55 million in HYPE through native staking within two weeks, according to Arkham on-chain data cited by etf.com.
The Aave DAO voted to make its buyback program permanent at $50 million per year, funded by protocol revenue running at $100-120 million annualized, according to The Defiant. The Aave Chan Initiative (ACI) proposal passed with 100% approval. Weekly buybacks range from $250,000 to $1.75 million depending on market conditions and protocol profitability.
An April 2026 governance vote redirected 100% of protocol revenue to AAVE holders, per CryptoDaily. The program has already retired more than 94,000 AAVE from circulation. The corporate entity behind Aave — Avara (formerly Aave Companies) — retains separate equity value, but the token-level value accrual mechanism is now formalized.
Uniswap governance activated its fee switch in December 2025, passing with 99.9% support (125 million tokens for, 742 against), per CoinDesk. The protocol burned 100 million UNI (~$596 million at the time) from the treasury and began routing approximately 17% of swap fees toward UNI buyback-and-burn.
Through Q1 2026, Uniswap burned over $5.5 million worth of UNI, implying an annualized pace of ~$34 million, according to CryptoDaily. Q1 2026 trading volume reached $231 billion across the protocol, with Unichain alone processing close to $12 billion monthly. The fee switch expanded to eight L2 networks following a March 2026 vote, per KuCoin.
The critical caveat: Uniswap Labs, the corporate entity, continues to collect its own 0.15% interface fee on certain front-end transactions — a revenue stream that accrues to shareholders, not UNI holders.
Jupiter allocated roughly half its protocol fee revenue — over $70 million — to JUP buybacks in 2025, according to Crypto.news. The token still fell to $0.20-$0.22 by early January 2026, an 89% decline from its peak.
The structural problem: monthly unlocks of 53 million JUP through June 2026 expanded circulating supply by approximately 150% since launch. The buyback volume covered only about 6% of unlocked tokens. Solana co-founder Anatoly Yakovenko commented that "in markets with heavy emissions, short-term buybacks do not reset how sellers price risk," per BeInCrypto.
Jupiter has since pivoted: the DAO cut its 2026 airdrop from 700 million to 200 million JUP and proposed a "Going Green" initiative targeting net-zero token emissions for the remainder of 2026, per SolanaFloor.
Helium founder Amir Haleem suspended HNT buybacks on January 2, 2026, stating that the market showed no interest in projects repurchasing their own tokens, per Phemex. Despite $3.4 million in monthly revenue, the team concluded funds would be better deployed toward business growth. Haleem subsequently resigned as CEO as HNT extended its 96% decline from all-time highs, per Protos.
The Helium case illustrates a broader point: buybacks without structural demand-side catalysts (staking sinks, governance utility, composability) amount to open-market purchases with no lasting impact on float dynamics.
Pendle retired its vePENDLE governance model in January 2026, replacing multi-year lockups with sPENDLE — a liquid staking token carrying a 14-day withdrawal period, per The Block. The protocol now allocates up to 80% of revenue toward PENDLE buybacks from the open market, with proceeds distributed to sPENDLE holders.
The transition also introduced a ~30% reduction in overall emissions and a 5% fee for instant redemption, per CoinDesk. Existing vePENDLE balances converted to boosted sPENDLE with multipliers up to 4x based on remaining lock duration. In 2025, vePENDLE holders earned over $37 million in fees, but the complexity of the voting system concentrated rewards among sophisticated users. The sPENDLE model democratizes access while retaining revenue-funded buyback mechanics.
CoW DAO proposed a three-pronged value return framework in May 2026: treasury token burns of 60-85 million COW, flexible buybacks scaling up to 100% of weekly protocol revenue under specific market conditions, and a restructured solver bond system requiring only 20% token commitment (down from prior levels), per CryptoTimes. The trial program runs through December 2026. This replaces the fixed buyback model in place since April 2024 with a dynamic system tied to token price, ETH conditions, and protocol profitability.
Treehouse, a DeFi protocol with $294 million in deposits, launched a buyback program committing 50% of Market Efficiency Yield fees from its tETH product to open-market TREE acquisitions, per DL News. The governance proposal passed with 99.59% support. Purchased TREE tokens are held in a DAO-controlled multi-sig as strategic reserves. The token price doubled following the announcement, though it remains 35% below its August peak.
Yield Basis, developed by Curve Finance founder Michael Egorov, activated its fee switch in December 2025, distributing 17.55 BTC (~$1.62 million) to veYB holders from $130 million in Bitcoin deposits, per DL News. The model bypasses token buybacks entirely, distributing protocol revenue directly in BTC — a structural choice that eliminates the intermediary step of buying back native tokens and instead delivers yield in a widely-held asset.
Ethena's fee switch, with benchmarks met in September 2025 and activation targeted for Q1 2026, directs 10-20% of protocol revenue to sENA stakers, per Cryptopolitan. The protocol generated $57 million in December 2025 alone and exceeded $500 million in cumulative revenue, per Tokenomics.com. An $890 million buyback program supplements the staking yield. The dual mechanism — direct fee sharing plus buyback support — represents the most capital-intensive value return system in DeFi outside of Hyperliquid.
The buyback landscape reveals a clear hierarchy of value accrual effectiveness:
| Protocol | Mechanism | Annual Scale | Emissions Offset? | Corporate Equity Separate? | |---|---|---|---|---| | Hyperliquid | Auto buyback (97% of fees) | ~$365M+ | No emissions | No VC, no equity | | Aave | Permanent buyback | $50M | Modest emissions | Yes (Avara) | | Uniswap | Fee switch + burn | ~$34M (annualized) | Treasury burn of 100M UNI | Yes (Uniswap Labs) | | Optimism | 50% revenue buyback | ~$8M est. | Heavy OP emissions | Yes (Optimism Foundation) | | Pendle | 80% revenue buyback | TBD (new system) | ~30% emissions cut | Yes (Pendle team) | | Ethena | Fee switch + $890M buyback | $57M/mo revenue | Active emissions | Yes (Ethena Foundation) | | Jupiter | Paused/pivoted | $0 (halted) | 150% supply increase | Yes (Jupiter team) | | Helium | Suspended | $0 (suspended) | Ongoing HNT emissions | Yes (Nova Labs) |
The pattern is clear: buybacks succeed when they are (1) funded by sustainable fee revenue, (2) executed against a backdrop of controlled or declining emissions, and (3) structurally embedded in the protocol rather than discretionary treasury decisions.
Protocols with separate corporate entities (Uniswap Labs, Avara, Nova Labs, Optimism Foundation) present a persistent governance tension: token holders fund buybacks from protocol revenue while the corporate entity may capture value through equity, interface fees, or service contracts. This dual-track value accrual remains under-analyzed relative to its importance.
The 2025-2026 buyback wave marks a structural shift in how DeFi protocols distribute value, but the data shows that mechanism design matters far more than dollar amounts. Hyperliquid's $1.3 billion in cumulative buybacks works because it operates without venture allocation, without emissions dilution, and with 97% of revenue committed. Jupiter's $70 million program failed because emissions dilution ran at 25x the buyback rate.
The protocols that will define the next phase of DeFi value accrual are those combining fee-funded buybacks with emissions reduction (Pendle's 30% cut), direct asset distribution (Yield Basis's BTC payouts), and governance-embedded automation. Discretionary, treasury-funded buybacks executed against heavy token unlocks — the Helium and Jupiter model — are demonstrably ineffective.
For token holders evaluating buyback programs, the critical variables are not the size of the buyback but the ratio of buyback demand to net new supply, the separation (or lack thereof) between corporate equity and token economics, and whether the mechanism is automated and transparent or discretionary and opaque.