Skip to content
Puffer (pufETH) logo

Puffer (pufETH) Review

Permissionless bonded operators and a priced instant exit

of 10

Puffer (pufETH) scores 7.8 out of 10 in liquid staking, up from 6.6 after its repository's six-firm audit list was counted. Open operator entry, GPL-3.0 code and two priced exits lead the score; the unpublished protocol fee holds it back. Based on primary sources checked on 1 November 2026.

Overall score

7.8 / 10 (3, 9, 9, 10, 8), up from 6.6

Protocol fee

Stated to exist; percentage not published

Withdrawals

To WETH: instant for 1%, or no fee in ~14 days (min 0.01 pufETH)

Operators

Permissionless; 2 ETH bond plus at least 28 validator tickets

Audits

BlockSec, SlowMist, Quantstamp, Trail of Bits, Nethermind, Creed; Immunefi

Licence

GPL-3.0 (PufferFinance/pufETH)

Scores

Commission clarity
3.0
Exit & redemption
9.0
Operator decentralisation
9.0
Licence & code openness
10.0
Audit disclosure
8.0

Each criterion carries equal weight. The overall score is the average of these 5 scores — (3.0 + 9.0 + 9.0 + 10.0 + 8.0) ÷ 5 = 7.8.

Pros

  • Operator decentralisation 9: permissionless entry with a 2 ETH bond and 28 validator tickets
  • Exit & redemption 9: instant to WETH for 1%, or no fee in around 14 days
  • Licence & code openness 10: GPL-3.0 on the pufETH repository
  • Audit disclosure 8: six named firms plus an Immunefi Boost, reports linked in the README

Cons

  • Commission clarity 3: a protocol fee exists but no percentage is published
  • Audit list lives only in a repository README; no report is dated after April 2024
  • Withdrawals pay WETH, and instant capacity depends on unpublished vault liquidity
  • Restaking through EigenLayer adds slashing exposure the five criteria do not score

pufETH is Puffer's staking token for Ethereum. Deposited ETH funds validators that are also restaked through EigenLayer, and each pufETH becomes worth more or less ETH as rewards and penalties accrue. In the liquid staking ranking it now scores 7.8 out of 10 on five equally weighted criteria, up from 6.6. The July review said no pufETH audit was named. The pufETH repository in fact lists audits by six firms, with linked reports, and that finding drives most of the change. The commission rate is still unpublished, and it remains the one low mark.

How Puffer pufETH scores in liquid staking

Each criterion is marked 0 to 10 and carries the same 20% weight. The basis is desk research of primary sources on 1 November 2026, not a test deposit or withdrawal.

CriterionFinding (checked 1 Nov 2026)Mark
Commission clarityThe rewards documentation says the treasury "is allocated a portion as a protocol fee" and that Guardians and restaking operators are paid from rewards. No percentage is published for any of them.3
Exit & redemptionTwo routes to WETH in the Puffer app. A 1-step withdrawal is instant for a 1% fee paid to the treasury, funded from the vault's liquidity. A 2-step withdrawal has no fee and takes around 14 days, often less; the minimum is 0.01 pufETH.9
Operator decentralisationThe requirements page says "anyone can become a Puffer NoOp because our protocol is permissionless". Operators lock a 2 ETH bond and at least 28 validator tickets per validator.9
Licence & code opennessThe PufferFinance/pufETH repository carries GPL-3.0.10
Audit disclosureThe repository README lists BlockSec, SlowMist, Quantstamp, Trail of Bits (March 2024), Nethermind and Creed (April 2024 file date), plus an Immunefi Boost review, across two versions. All but SlowMist's second review are linked.8

The arithmetic: (3 + 9 + 9 + 10 + 8) / 5 = 39 / 5 = 7.8.

What changed: audit disclosure rose from 3 to 8. The July review found only a governance-token audit and missed the audit list in the pufETH repository. Exit and redemption rose from 8 to 9 to match the standard applied to Ankr. Like Ankr, Puffer publishes an instant route with a stated price and a standard route with no fee. Commission, operators and licence are unchanged.

Strongest criteria: open operators, open code and a priced exit

Operator decentralisation (9). Entry is open to anyone who posts the bond and buys tickets, rather than to a curated list. That is the rubric's highest operator standard, shared with Stader's permissionless pool. Each validator still needs 32 ETH, of which the operator puts up 2 ETH, about 6%. The bond is forfeited if the validator is slashed, which puts the operator's own money at risk first. Puffer also says it caps its share of Ethereum's validator set at 22%. The mark stops short of 10 because the restaking layer is curated: Puffer says it restricts which EigenLayer services and restaking operators are allowed.

Licence & code openness (10). GPL-3.0 is a recognised open-source licence, the same standard that gives Stader, Bedrock, Mantle and Origin 10. One caveat sits outside the mark: the pufETH repository was last updated in June 2024.

Exit & redemption (9). A holder can choose speed or cost and knows the price of each. On a 10 ETH withdrawal, the instant route costs 0.1 ETH plus gas, and the standard route costs only gas but can take about two weeks. Both routes pay WETH, not native ETH. As with Ankr's flash pool, instant capacity depends on how much liquidity the vault holds, and no figure for it is published, so the mark is not a 10.

Weakest criteria: what is missing and what was measured

Commission clarity (3). This is a disclosure gap, and the single most consequential one for a staking token. The documentation names three groups paid from rewards: the treasury, Guardians and restaking operators. It gives no rate for any of them. Node operators keep 100% of consensus and execution rewards for the days their tickets cover, and ticket payments flow to pufETH holders, so the fee structure differs from a simple commission. That makes a published rate more important, not less. Without it, a holder cannot work out the share of restaking rewards that reaches them. Our guide to staking returns after costs shows why the fee figure matters.

Audit disclosure (8). Six firms are named and the reports can be read, which meets the category's definition. The mark is one below the 9s of Stader, Mantle, Origin and Liquid Collective for three stated reasons. The list exists only in a repository README; the documentation site has no audit page. One listed review, SlowMist v2, has no link. No report is dated after April 2024.

What the score does not cover

  • Restaking risk. Puffer describes pufETH as a native liquid restaking token. Its backing can fall if a restaking operator is slashed while running EigenLayer services, on top of Ethereum penalties. No criterion here scores that exposure; the liquid restaking ratings do.
  • Upgrades. A team 3-of-6 multisig acts through a 7-day timelock, but a community 3-of-6 multisig can veto it or bypass the timelock. Governance is not scored.
  • Ticket pricing. Guardians currently post the validator ticket price. How well that price tracks validator revenue was not measured.
  • Yield. Neither the pufETH rate nor its price is a criterion.

Puffer vs Mantle and Origin Ether

On 1 November 2026 marks, pufETH sits between Mantle and Origin Ether in the liquid staking ranking. Both were re-checked the same day.

CriterionMantle (mETH)Puffer (pufETH)Origin Ether (OETH)
Commission clarity838
Exit & redemption898
Operator decentralisation693
Licence & code openness101010
Audit disclosure989
Overall8.27.87.6

Puffer's profile is the reverse of its neighbours'. It leads both on operators and exit, and trails both by five points on commission. Mantle finishes 0.4 ahead because its stated fee outweighs Puffer's open operator set. Puffer finishes 0.2 ahead of Origin Ether, whose validators run on one provider. If Puffer published its protocol fee, commission would be re-checked, and that alone could move it above both.

Who this rating suits

The score is most useful to a holder who wants validators run by an open, bonded operator set, open-source code and a choice between a priced instant exit and a free standard one. It is less suited to someone who needs to know the protocol's cut before staking, or who wants plain staking without a restaking layer. The rating describes published terms on one date. It is not a recommendation to stake and does not assess any individual's circumstances.

How it compares

Sources

Scored from published primary sources rather than hands-on use. Every figure above comes from one of the pages below, on the date shown. Manufacturers change prices and specifications without notice — if something here no longer matches the source, tell us and we will correct it.

  1. 1.GitHub: PufferFinance/pufETH README (audits) and licence (GPL-3.0) — checked
  2. 2.Puffer docs: rewards (fee wording) — checked
  3. 3.Puffer docs: withdraw — checked
  4. 4.Puffer docs: node operator requirements — checked
  5. 5.Puffer docs: native liquid restaking token — checked
  6. 6.Puffer docs: validator tickets — checked
  7. 7.Puffer docs: multisigs — checked
  8. 8.Trail of Bits: Puffer Finance security review (March 2024) — checked

Frequently asked questions

Why does Puffer score 7.8 in liquid staking?

Its marks are 3 for commission clarity, 9 for exit, 9 for operator decentralisation, 10 for licence and 8 for audits. They total 39, and 39 divided by 5 is 7.8. Each criterion counts equally.

What changed since the previous score of 6.6?

Audits rose from 3 to 8 because the pufETH repository lists six audit firms with linked reports, which the July review missed. Exit rose from 8 to 9 to match Ankr, which also publishes a priced instant route and a free standard one.

What holds the Puffer score back?

The protocol fee. Puffer's documentation says the treasury takes a portion of rewards but gives no percentage, so commission clarity scores 3. Publishing the rate is the change that would most affect the score.

Was pufETH tested with a real deposit or withdrawal?

No. The rating is desk research of Puffer's documentation and GitHub repositories. The 14-day standard withdrawal is Puffer's stated estimate, not a measured wait.

How does Puffer compare with Mantle and Origin Ether?

Mantle scores 8.2 and Origin Ether 7.6. Puffer leads both on operators and exit, and is level on licence. It trails both by five points on commission and one point on audits.

Nothing here is financial advice. Editorial policy · How we score · How we're funded.