# Marinade Q2 2026 quarterly report

> Stake fell 4.5% in SOL and dollar revenue fell a third as the market pulled back. The stake auction cleared above the network benchmark in every epoch of the quarter, before Marinade's fee.

Canonical HTML: https://app.marinade.finance/learn/marinade-q2-2026-quarterly-report/

*April 1 to June 30, 2026 · Solana epochs 950–995 · all figures as of June 30, 2026.*

| Headline | Value | Change |
|---|---|---|
| SOL staked, end of Q2 | 6.55M SOL | −4.5% QoQ |
| Revenue, Q2 | $795K | −32.6% QoQ |
| Protected Staking Rewards (PSR) paid by bonds | 85.5% | of 337.2 SOL paid to stakers |
| Auction yield vs benchmark, before Marinade's fee | +0.32% | ahead in 46 of 46 epochs |
| Institutional stake (Select) | 1.21M SOL | +32.3% QoQ |

> [!NOTE] New to Marinade? What the protocol does, in four lines
>
> - **The products.** Marinade helps retail users, DAOs and institutions manage and optimize their staked SOL. **mSOL** is a token you hold and trade while the SOL behind it stays staked; **Native** stakes from your own wallet without giving up custody; **Select** serves institutions, ETFs and clients of qualified custodians; **Customized Rewards** pays rewards in a token of your choice.
> - **The auction.** Validators bid for Marinade's stake in the Stake Auction Marketplace (SAM), backing each bid with collateral called a bond. Their bids raise what stakers earn; Marinade keeps a share as revenue.
> - **The benchmark.** The Solana Staking Index (SSI) measures what plain, neutral staking on Solana pays. It is the yardstick used throughout.
> - **The token.** MNDE is the governance token of the Marinade DAO, which owns the protocol; proposals are called MIPs and are voted on-chain.

## The quarter in five lines

1. **The decline was in mSOL and Native.** Institutional stake grew 32% and is now close to a fifth of Marinade's total.
2. **The auction beat the benchmark in all 46 epochs**, 6.42% against the index's 6.10%, before Marinade's fee; Native stakers earned 0.21 points below the index on average.
3. **Most of the revenue fall is the SOL price.** Measured in SOL, revenue held near 3,300 a month.
4. **MIP-22 passed unanimously.** From Q3, 10% of auction fees buys MNDE for MNDE stakers who vote.
5. **Shipped.** Benchmark-linked dynamic yield, staking through Anchorage custody, a rebuilt PSR dashboard, and the audited mSOL upgrade that reaches mainnet in Q3.

## Letter from the core contributor

Q2 was a hard quarter, so I'll start with that. We wanted to grow stake. It went down a little. SOL dropped by about a third in the middle of the quarter, a lot of leverage left the market, and our revenue in dollars fell with the price.

Some things went well. Select kept growing. The mSOL upgrade passed its audit, and the DAO voted for MIP-22, so part of protocol revenue now buys MNDE for the people who stake it and vote.

The most useful feedback came from our own team. In June we asked everyone what was slowing us down. The top answer was a simple question: who are we building for? People also told us we start too many things and that some parts of Marinade got too complicated. I agree with them. We spent too long building in our own room. So this quarter each of us is talking to at least one person who uses Marinade.

We also got clearer on what matters. Revenue counts for more than size, and in SOL our revenue fell much less than in dollars, even as stake went down. A slightly higher APY rarely makes anyone move their SOL. People want staking that is simple from the first deposit to the last withdrawal, and a way to use staked SOL without selling it.

For Q3 we kept the list short and focused on finishing what we started. Better yield for stakers. A way to borrow against staked SOL so you don't have to unstake. Staking rewards reports that are easy to get. And better internal tools, so the team spends less time hunting for answers.

The goal hasn't changed. We want every SOL working safely and earning a good rate, whether it sits in a phone wallet or an ETF. It doesn't matter much to us if people stake through mSOL, Native or a validator they chose themselves. A safe, fast Solana is good for everyone on it.

The market is quiet right now, which is a good time to get ready. If you stake through Marinade and something bugs you, tell me. My DMs are open.

repe, core contributor

## mSOL and Native fell, Select grew

In dollars the fall is steeper, −13.1%, because SOL fell too. Select's share of Marinade's total stake rose from 13.4% to 18.5%.

| SOL staked | Q3 2025 | Q4 2025 | Q1 2026 | Q2 2026 | QoQ |
|---|---|---|---|---|---|
| Total | 10,318,729 | 9,359,013 | 6,862,729 | 6,552,849 | −4.5% |
| Liquid (mSOL) | 4,493,393 | 3,444,359 | 2,813,440 | 2,437,004 | −13.4% |
| Native | 5,037,772 | 3,128,243 | 2,853,413 | 2,559,372 | −10.3% |
| Select (institutional) | 897,167 | 2,747,669 | 917,145 | 1,213,840 | +32.3% |
| Customized Rewards | 0 | 0 | 258,876 | 305,009 | +17.8% |

Source: Marinade API (tlv/history), end-of-quarter snapshots. The mSOL row is net of pending unstake tickets (30,109 SOL at June 30); adding those and the instant-unstake liquidity pool (7,515 SOL) reconciles the rows to the total.

![Stake rose from 6.86M SOL to 7.34M by the end of April, then fell to 7.09M at the end of May and 6.55M on June 30.](https://public.marinade.finance/learn/marinade-q2-2026-quarterly-report/tvl-chart.webp)

Three market forces weighed on mSOL and Native. SOL fell through the quarter, shrinking leveraged positions across the network. In June, lending rates on Solana's largest credit markets rose above staking yields, unwinding leveraged positions that borrow against liquid staking tokens. And the category itself shrank: stake in Solana stake pools fell 2.3% between epochs 950 and 995 while total network stake grew 1.2%, cutting the pooled share of all staked SOL from 28.1% to about 27%.

Select grew 32%, most of it in the last week of May, and Customized Rewards grew 17.8%, mostly in April and early May.

## Revenue fell a third in dollars, far less in SOL

Marinade earns its fees in SOL, and this report measures them in SOL first. Q2 revenue was **9,897 SOL**, steady at about 3,300 SOL a month. In dollars it was **$795,154**, a third below Q1.

The dollar fall is mostly price. The rest is a smaller average stake and a lower Marinade fee, which Marinade cut from late May to hold Native stakers at or just below the benchmark until June 22. In SOL the fall was about half as steep. Gross bid charges, what validators paid into the auction, fell 16.9% in SOL, against a 32.6% fall in dollar revenue.

| Revenue | Q1 2026 | Q2 2026 | QoQ |
|---|---|---|---|
| Revenue (USD) | $1,179,173 | $795,154 | −32.6% |
| Revenue (SOL) | n/a | 9,897 SOL |   |
| Validator bond settlements (SAM and Select fees) | n/a | 7,611 SOL |   |
| mSOL unstake fees | n/a | 1,785 SOL |   |
| Native delayed-unstake fees | n/a | 501 SOL |   |
| **What moved it** |   |   |   |
| Average stake over the quarter | 7.75M SOL | 7.01M SOL | −9.7% |
| Average SOL price | $102.81 | $80.58 | −21.6% |
| SAM bid charges, gross (on-chain) | 11,990 SOL | 9,964 SOL | −16.9% |
| **Staker protection (PSR)** |   |   |   |
| Cover paid to stakers (downtime, commission increases) | 157.6 SOL | 337.2 SOL | +113.9% |
| of which paid by validators' bonds | 84.3% | 85.5% | +1.2 pts |

Protected Staking Rewards made stakers whole for 34 downtime and commission-increase events across 19 validators, and validators' bonds paid most of the cost.

Governance decides where revenue goes. **MIP-17**, passed December 26, 2025, suspended the earlier split that sent 50% of fees to MNDE buybacks. Under MIP-15 and MIP-21 protocol fees go to the DAO, so all Q2 revenue accrued to the DAO.

## Yield against the benchmark

Marinade's auction cleared above the Solana Staking Index in **all 46 epochs** of the quarter, averaging **6.42%** against the index's 6.10%. That is the auction's yield before Marinade's own fee; what stakers earn is that yield minus the fee.

Over the quarter Native stakers earned 6.08% against the index's 6.29%, 0.21 points below it. These two figures use the APY API's own basis, which compounds per epoch and puts the index about 0.2 points higher. mSOL holders earned about 6.1%, measured from the mSOL price. The gap moved with the fee. Until late May the fee was static and Native ran about 0.27 points below the index. From May 23 the dynamic fee, completed on May 19, held Native close to the index, 0.03 points below it on average. From June 22 Marinade set the fee to secure a minimum SOL revenue instead, and Native ran about 0.45 points below the index for the quarter's last five epochs. Widening the auction's margin takes more validators bidding with more collateral behind them, which is the Q3 priority.

![The shaded area is the auction's margin over the index, before Marinade's fee. The widest gap was +1.10% in epoch 951.](https://public.marinade.finance/learn/marinade-q2-2026-quarterly-report/apy-chart.webp)

The index measures neutral staking: inflation plus block rewards, maximal extractable value (MEV) excluded. The auction's clearing yield is measured before Marinade's fee, and part of the spread is structural, since SAM captures MEV that the index excludes by construction. Both series are annualized on each epoch's measured duration, the basis Marinade uses internally; that choice moves the levels about 4 basis points and leaves the spread unchanged.

## MNDE, treasury and value accrual

**MIP-22**, the quarter's only MNDE-holder vote, was approved on Realms on June 29 with a unanimous Yes at 25.9M vote weight. From Q3, 10% of auction fees buys MNDE for MNDE stakers, and 10M MNDE seeds an MNDE/mSOL pool. No buybacks ran during Q2, since MIP-17 had suspended the earlier program and MIP-22 is its successor.

| MNDE snapshot (Jun 30) | Value |
|---|---|
| Price | $0.0187 (+6.9% QoQ) |
| Total supply (on-chain) | 700.0M |
| Circulating, measured (total minus DAO reserve) | 510.3M |
| Implied market cap at measured circulating | ≈$9.6M |
| MNDE staked in governance | 83.35M (16.3% of circulating) |
| MNDE staking accounts | 15,179 |

> [!NOTE] For MNDE holders: how to qualify for the MIP-22 revenue share
>
> - **Stake and vote.** Eligible wallets have MNDE staked in Marinade's Realms governance *and* have voted on at least one community proposal in the twelve months up to the end of the quarter; council votes do not count.
> - **What you receive.** MNDE bought on the open market with 10% of auction fees, shared pro rata to each wallet's average voting weight (veMNDE) over the quarter and distributed after each calendar quarter ends. Distributed MNDE can be claimed and restaked.
> - **When it starts.** The first buyback cycle begins with the first full quarter after on-chain implementation, and the first distribution follows once that quarter ends. Amounts depend on auction fees, the SOL price and total MNDE staked; actuals are published each quarter.

The DAO's balance sheet is on-chain and readable in any explorer. At June 30 it held **344.5M MNDE** across three accounts and one protocol-owned liquidity position, $6.45M at that day's price, against roughly **$95K** of non-MNDE liquid assets. Most of the treasury is held in MNDE, so its realizable value depends on market depth as well as on the quoted price.

### Account-level balances (on-chain, June 30)

| Account | Role | Holdings | USD (spot) |
|---|---|---|---|
| B56RWQGf…25YtvG | DAO reserve treasury (Realms; held since inception, non-circulating) | 189,716,233 MNDE | $3.55M |
| J5BEceL5…FfCzhs | Council budget vault (Realms; funds DAO-authorized spending) | 113,104,161 MNDE | $2.12M |
| 89Srbjbu…CeArBgB | DAO operations wallet | 30,696,570 MNDE + 877 mSOL + 643 USDC | $0.58M + $0.09M |
| FSjg4bkt…EXWQd35 | Orca MNDE/mSOL position owned by the operations wallet (protocol-owned liquidity, 1% fee tier; MNDE only, deposited December 7, 2025) | 10,956,000 MNDE | $0.21M |

Balances read on-chain as of June 30, 2026, valued at that day's prices (SOL $75.01, MNDE $0.0187). **After the quarter:** on July 7, 2026 the operations wallet closed the Orca position above and placed 10M MNDE in a new position in the 0.3% MNDE/mSOL pool named by MIP-22. MIP-22 buybacks started in July. Both appear in the Q3 report.

## Shipped in Q2

Most of the quarter's engineering went into MIP-19, the work of pricing the auction against the benchmark. The rest went to institutional access and to preparing the mSOL contract upgrade.

| Date | Release | What it does |
|---|---|---|
| Apr 21 | Bond Risk Reduction Mechanism | Under-bonded SAM validators lose stake and pay a fee to stakers |
| Apr 23 | Anchorage Digital integration | Marinade staking through a federally chartered custodian |
| Apr 30 | Priority activation | Auction winners can pay from bond to activate stake faster |
| May 19 | Dynamic SSI-based yield | mSOL and Native APY unified around the benchmark (MIP-19 Phase V.b) |
| May 27–28 | Earn pages + Unstake-to-USDC | One-step exit from staked SOL into the USDC Vault (on Kamino, strategy by RockawayX) |
| Jun 23 | PSR Dashboard V2 | Rebuilt with bond health, fee breakdowns, PSR estimates |
| Jun 24–30 | mSOL contract upgrade prepared | Neodyme audit passed, code merged, SDK v6.0.0 released; mainnet rollout in Q3 |
| Jun 30 | SOC 2 Type II report renewed | Second-year report issued by BDO, covering the June 2025 to May 2026 audit period |

## Where Marinade stands

Measured on-chain at epoch 995, the epoch spanning quarter end, Marinade's pools held 6.49M SOL of delegated stake, down 4.5% over the quarter. Among commercial staking products, that puts Marinade behind Binance (10.0M SOL) and Jito (9.9M), and ahead of Jupiter (5.3M), DoubleZero (5.2M) and Drift (2.8M). Delegated stake is measured at an epoch rather than at quarter end, so it differs slightly from the 6.55M total above.

Stake declined across most of the sector. Marinade's delegated stake fell less than Jito's and DoubleZero's, while Binance and Jupiter grew. Dollar revenue fell 32.6% here, against 45% at Jito and 39.7% at Sanctum; mSOL's SOL value fell 13.4% where jitoSOL's fell 20%. Over the same period Sanctum's aggregate TVL reached an all-time high of 16.6M SOL; that figure aggregates many issuers' tokens rather than one staking operation.

Marinade offers liquid, native and institutional staking on one validator set, priced by the same auction and protected by the same bonds. The **Canary Marinade Solana ETF (Nasdaq: SOLC)**, live since November 2025, accrues its staking rewards to NAV through Marinade, with BitGo as qualified custodian. April's **Anchorage Digital** integration lets allocators stake through Marinade without leaving federally chartered custody, and in June Marinade decided to expand the Select validator set for Q3.

Ranking covers commercial staking products. Three larger pools are excluded as delegation programs rather than products: the Solana Foundation's (24.2M SOL), the Firedancer client-adoption program (7.5M), and one unattributed 24.0M pool spread evenly across six validators. Sanctum runs many separate pools, one per liquid staking token, rather than one staking operation and is not comparable on this basis. Peer figures come from Jito's and Sanctum's own published Q2 2026 reporting; no per-token realized yield series is published by all three, so this report does not rank Marinade on realized yield.

Data: [Trillium](https://trillium.so).

## Priorities for Q3

- **Better yield for stakers.** Widen the auction's margin over the index with more validators and more bonded capacity.
- **Borrow against staked SOL.** Let stakers borrow from the Marinade app without unstaking.
- **Staking rewards reports.** Make per-account rewards reports easy to get.
- **Better internal tools.** So the team spends less time hunting for answers.
- **Already in flight.** The Neodyme-audited mSOL contract upgrade to mainnet (the first liquid-staking program upgrade since 2023), the start of MIP-22 buybacks and the first distribution to eligible MNDE stakers, and the expanded Select validator set.