Puffer Finance
About
Puffer Finance is a decentralized native liquid restaking protocol (nLRP) built on EigenLayer that allows anyone to run Ethereum PoS validators with just 2 ETH while supercharging rewards through restaking. It issues pufETH, a native Liquid Restaking Token (nLRT) that accrues both traditional Ethereum PoS rewards and EigenLayer restaking rewards simultaneously, and is fully composable across DeFi. The protocol also develops UniFi, a based rollup (L2) offering instant L1 withdrawals, alongside the UniFi Preconf AVS for sub-second preconfirmation services on Ethereum.
Where Does Yield Come From?
pufETH holders earn yield from three separate sources working together. Here is how each one works.
1. Validator Tickets (VTs)
Validator Tickets are tokens that give a Node Operator the right to run an Ethereum validator (funded by stakers) for one day. To start a validator, the operator must lock up 2 ETH in pufETH as collateral and hold at least 28 VTs.
The price of a VT is set by oracles based on what a validator is expected to earn in one day (from both consensus and execution rewards), minus a small discount. The price updates every 12 hours, or sooner if the market moves a lot.
When someone buys a VT, the ETH they pay gets split: part goes to the Guardians (for fees and gas costs), part goes to the protocol's treasury, and the rest goes into the PufferVault. That deposit directly increases the amount of ETH backing each pufETH token. This means pufETH holders earn their proof-of-stake rewards upfront — right when the VT is minted — instead of waiting for a specific validator to perform over time.
The Node Operators keep 100% of their validator's execution rewards (including MEV, which they receive instantly through their own fee address) and 100% of consensus rewards.
2. Native Restaking via PufferModules
Node Operators run their validators inside special contracts called PufferModules. Each module controls an EigenPod, which acts like a single restaking pool made up of many validators. The pooled ETH is then used as collateral to secure services on EigenLayer (called AVSs).
The DAO selects Restaking Operators (ReOps) to perform duties for those services. These operators earn a share of the service fees, and the rest flows back to the protocol and pufETH holders. Operators whose validators are inside a PufferModule automatically earn extra restaking rewards on top of their full proof-of-stake rewards.
During this early stage, both the operators and the services they secure are carefully chosen by the DAO to reduce the risk of slashing (penalties). The DAO also decides which modules to create and which services they connect to.
3. pufETH as a Reward-Bearing Token
pufETH is built like a "cToken" — a token whose value in ETH rises over time. As the protocol collects revenue (from VT minting and restaking rewards) and subtracts any penalties or slashing, pufETH's exchange rate against ETH goes up.
Stakers earn rewards from both proof-of-stake and restaking just by holding pufETH. There is no need to separately restake a liquid staking token into another protocol.
The protocol also sets a self-imposed cap: it will not let pufETH control more than 22% of all Ethereum validators, to keep the network neutral and decentralized. The $PUFFER governance token (locked as vlPUFFER) gives holders more voting power and influence over protocol decisions, but it is not itself a source of yield for pufETH's core value.
Persons
Amir (amiru.eth)
Co-Founder of Puffer / Co Founder at Puffer Labs
Jason Vranek
Co-Founder and CTO of Puffer Labs
Audits
| Audit / Date | Findings | Verdict |
|---|---|---|
Trail of Bits25-03-2024 |
| The audit found no critical issues and the single high-severity finding (DOS on withdrawals) was resolved; the two medium-severity findings that remain unresolved were acknowledged by the team as expected behavior or planned deprecation, so the contracts present manageable residual risk. |
Nethermind12-04-202416-04-2024 |
| The audit found no critical vulnerabilities and resolved the single high-severity front-running issue prior to publication, indicating the protocol's core security posture is sound. Residual acknowledged items (e.g., hash collision mitigation, slippage protection) represent accepted design trade-offs that should be monitored in future upgrades. |
SlowMist09-04-2024 - 22-04-2024 |
| The audit found no critical or high severity vulnerabilities; one medium and one low finding were acknowledged by the team and one low finding was fixed, indicating a reasonable security posture for Phase 2 prior to mainnet deployment. |
BlockSec23-04-2024 |
| The report shows one high-severity and two medium-severity issues were addressed (Fixed) or acknowledged, with no unmitigated critical vulnerabilities; residual concerns include signature collision risks (EIP-712 planned) and centralization through multi-sig control, which are typical for early-stage protocols but warrant monitoring as the protocol matures. |
BlockSec29-01-2024 |
| The audit found no critical or high-risk vulnerabilities in the pufETH contracts; the single low-severity issue and all recommendations were addressed by the project team. The residual notes highlight design considerations (MEV exposure, airdrop distribution) rather than exploitable flaws, making the contracts reasonably safe for their stated scope. |
SlowMist24-01-2024 - 29-01-2024 |
| The audit identified one medium and one low severity issue plus a process suggestion, all of which were resolved by the team. The findings do not raise residual safety concerns for the contracts as deployed. |
Quantstamp01-02-2024 - 07-02-2024 |
| The audit identified no high or critical findings, with the only medium-severity issue (permit/deposit fund mismatch) already fixed in deployed contracts; the residual low and informational items were largely acknowledged or mitigated by the team, making this a clean report for the pre-mainnet version of pufETH that does not yet participate in EigenLayer restaking/delegation. |
| The audit document was accessible but its contents could not be parsed into a usable format for analysis, so no determination of protocol safety can be made from this file alone. | |
Trail of Bits25-03-2024 |
| Trail of Bits identified 6 findings (1 High, 2 Medium, 2 Low, 1 Informational) in Puffer Finance's smart contracts, with the most critical being a denial-of-service vector for withdrawals that was subsequently resolved. Four issues were fixed, while two medium/low items were acknowledged as acceptable design trade-offs by the Puffer team, leaving residual risks around withdrawal edge cases and stETH rebasing. |
0xLuckyLuke15-01-2024 |
| One critical signature-validation gap was identified in the proof-of-reserve oracle, alongside several low-severity architectural and input-validation concerns; all findings were accompanied by clear remediation guidance, and no medium-severity issues were found, indicating a reasonably solid codebase with manageable pre-deployment fixes. |
Cantina Managed (Sujith Somraaj, Ladboy233)16-04-2025 - 19-04-2025 |
| The audit found no critical, high, or medium vulnerabilities, and the single low-risk finding was promptly fixed, indicating a solid security posture for the VaultV5 upgrade. |
Cantina (Windhustler, OxWeiss)04-02-2025 - 05-02-2025 |
| This sCARROT targeted review found no critical, high, or medium severity issues; all low and informational findings were addressed (three fixed, one acknowledged), indicating a clean security posture for the audited scope. |
BlockSec23-04-2024 |
| The high-severity finding was resolved in Version 2, but several medium and low issues remain confirmed and unaddressed, posing residual risks that the protocol should carefully evaluate before mainnet deployment. |
BlockSec23-04-2024 |
| The audit found one high-severity issue (fixed) and two medium-severity issues (one fixed, one confirmed), indicating that the most critical risks were addressed before deployment; residual design concerns around guardian message collisions and centralization remain noted by the auditor. |
BlockSec29-07-2024 |
| The single high-severity finding (incorrect migrate amount) was fixed, and all low-severity issues were either fixed or acknowledged, leaving no unresolved critical or high risks in the audited scope at the time of reporting. |
BlockSec03-09-2024 |
| The audit found no critical or high-severity vulnerabilities; all three medium-severity issues were fixed, and the two low-severity confirmed items are either accepted as unlikely or will be handled off-chain. The Fast Path Contracts present a reasonable security posture for production deployment given the resolved findings and reliance on a multisig for privileged operations. |
BlockSec24-09-2024 |
| BlockSec's audit identified three security issues (1 Medium, 2 Low) in Puffer's 2-step withdrawal contracts, all of which were fixed or acknowledged by the project; the residual low-risk sandwich attack vector was deemed economically impractical by the team. Overall, the contracts appear safe for the intended withdrawal functionality with no unaddressed high or critical risks. |
BlockSec25-09-2024 |
| The audit found no exploitable vulnerabilities in the Puffer token contract, with only low-severity recommendations and acknowledged centralization notes, indicating a clean security baseline for the audited scope. |
BlockSec27-11-2024 |
| The audit found no critical or high-severity vulnerabilities; the single Low-severity issue and all four recommendations were fixed by the team, meaning no unpatched security findings remain in the audited scope, though several design-level notes (centralization, exchange rate discrepancies) are acknowledged as residual considerations. |
BlockSec25-07-2025 |
| Three low-severity issues were found, one of which was fixed before the final report, with the remaining two acknowledged by the Puffer team as having negligible practical impact; the audit focused exclusively on the withdrawal fee rework changes and did not identify any critical or high-risk vulnerabilities. |
SlowMist24-01-2024 - 29-01-2024 |
| The audit identified one medium and one low severity finding, both of which were fixed before deployment, indicating the protocol addressed meaningful design risks. No critical or high vulnerabilities were found, supporting a reasonable security posture for the pufETH contracts at launch. |
SlowMist09-04-2024 - 22-04-2024 |
| The audit found no critical or high-severity vulnerabilities, with the single medium and low finding being either acknowledged or fixed by the team, and the code had not yet been deployed to mainnet, indicating manageable residual risk. |
SlowMist27-06-2024 - 28-06-2024 |
| The audit identified only low-severity and suggestion-level issues, all of which were either fixed or acknowledged with a clear scoping decision; no critical, high, or medium vulnerabilities were found, indicating the L2 staking contracts were in a solid security posture prior to deployment. |
SlowMist09-01-2025 - 10-01-2025 |
| The audit found one medium-risk excessive privilege issue (Governor can drain user underlying tokens) that was acknowledged but not resolved, keeping the overall conclusion at medium risk; all other findings were fixed or acknowledged, so the contract is reasonably secure assuming the privilege risk is managed through governance safeguards. |
SlowMist11-09-2025 - 15-09-2025 |
| The audit revealed one acknowledged medium-risk finding regarding excessive owner privilege, with no critical or high-severity vulnerabilities identified; the contract is otherwise sound from a security standpoint, though reliance on a single privileged owner remains the primary residual risk. |
SlowMist11-09-2025 - 15-09-2025 |
| The audit found no critical or high-severity vulnerabilities; the two low-risk findings were acknowledged by the team with existing multi-sig/timelock mitigations, and all suggestion-level findings were fixed. The Rewards contracts present a moderate residual risk from privileged roles, but this is partially addressed by the operational controls described. |
Quantstamp01-02-2024 - 07-02-2024 |
| The audit shows no critical or high-severity vulnerabilities; the single medium-severity finding (permit-vs-msg.sender fund usage) was fixed before deployment, and residual risks around swap functionality and privileged roles were acknowledged or mitigated by the team, making the contracts acceptably safe for their pre-mainnet staking release. |
| This Immunefi Boost audit report for Puffer Finance's pufETH v1 could not be analyzed because the PDF conversion yielded only garbled, unusable text — the actual findings, severity counts, and remediation details are not extractable from the document as processed. | |
0xLuckyLuke15-01-2024 |
| The single High-severity finding (unchecked oracle state update) poses a material risk and must be resolved before mainnet deployment, while the Low and Informational items indicate code-quality areas for improvement but no systemic architectural flaws. |
Creed03-06-2024 - 07-06-2024 |
| The review found no critical or high-risk issues; the two items identified were acknowledged or addressed, indicating the in-scope contracts were in a solid security posture at the time of audit. |
Nethermind12-04-202416-04-2024 |
| The audit uncovered one critical-adjacent High-severity front-running risk (fixed by restricting provisionNode to a paymaster, albeit at the cost of some centralization) and several Medium/Low issues, most of which were resolved; the codebase was assessed as high quality overall, with residual acknowledged risks around guardian message hashing and module selection fairness that warrant continued monitoring. |
Trail of Bits25-03-2024 |
| Four of six findings were fixed before publication, with the two remaining medium/low-risk issues acknowledged as acceptable design choices; the audit provides reasonable assurance for the reviewed contracts but notes that further fuzz testing and documentation of access controls are recommended. |
Cantina Managed (Gerard Persoon, Ladboy233)29-10-2025 - 30-10-2025 |
| The Carrot Vesting Upgrade audit found no critical or high-severity issues; the single medium and four low-severity findings were all acknowledged by Puffer Finance with no in-scope fixes, indicating the code was accepted as-is with documented residual risks. |
Backers
Puffer Finance has raised multiple funding rounds. A $5.5 million seed round was co-led by Lemniscap and Lightspeed Faction in August 2023, with participation from Brevan Howard Digital and Jump Crypto among others. A $650,000 pre-seed round preceded this. In January 2024, Binance Labs invested an undisclosed amount in Puffer. In April 2024, Puffer raised an $18 million Series A round led by Brevan Howard Digital and Electric Capital, with participation from Coinbase Ventures, Kraken Ventures, Lemniscap, Franklin Templeton, Avon Ventures (affiliated with FMR LLC/Fidelity's parent company), Mechanism Capital, Lightspeed Faction, Consensys, Animoca Brands, GSR, and others. Puffer also received a $120,000 grant from the Ethereum Foundation for its open-source Secure-Signer tool.
Institutional backers listed on the official Puffer website include: Brevan Howard Digital, Electric Capital, Lemniscap, Faction (Lightspeed Faction), Coinbase Ventures, Binance Labs, Franklin Templeton, F-Prime, Kraken Ventures, Consensys, Bankless Ventures, and Mechanism Capital. Angel investors listed include Sreeram Kannan (Founder of EigenLayer), John Zettler (Staking Product Lead at Coinbase), Sandeep Nailwal (Co-founder of Polygon), Andrew Kang (Partner at Mechanism Capital), Stephane (Co-founder of Flashbots), Chao Jijun (Founder of Benmo), DiscusFish (Co-founder of F2Pool & Cobo), Ted Lin (Web3 Angel and former CGO at Binance), YaoQi (Co-founder of AltLayer), Calvin (CSO at EigenLayer), Anton Buenavista (Founder of Pendle), and 0xCygaar (Co-founder of Frame). Advisors include Justin Drake (Ethereum Foundation researcher), DeFi Dad (Partner at 4RC), Mr. Block (Core Contributor at Curve), Allan Zhang (Founder of BayAreaWSB), Jason Chen (crypto researcher), Anthony Sassano (Founder of The Daily Gwei), Michael Egorov (Founder of Curve), and DiscusFish.
Legal
Legal form
Foundation (referred to as "Puffer Foundation" in official docs)
Status and notes
Copyright footer states "Copyright © 2022 - 2026 by Puffer Finance. All rights reserved." Official documentation mentions "the Puffer Foundation" was established to guide protocol governance and transition to a DAO. GitHub organization uses name "PufferFinance" with contact email [email protected]. No imprint, entity registration number, or jurisdiction disclosed on official sources.