Solana inflation: the schedule, where it stands, and what was voted
New SOL is printed every epoch and paid to stakers. The rule that sets how much has not changed since 2021, although two attempts to change it went to a vote.
Solana's network inflation rate was 3.556% per protocol year on November 16, 2026 (epoch 1,087, getInflationRate), which works out to about 3.37% of supply per calendar year because real slots run slower than the 200 ms the protocol counts in; it follows a fixed schedule read from getInflationGovernor: 8% at the start in February 2021, falling by 15% of itself each year to a 1.5% floor, about 5.3 protocol years away. A 2025 plan to tie issuance to staking (SIMD-0228) failed; a 2026 plan to double the rate of decline (SIMD-0550) is reported to have passed a stake vote but is not active on-chain.
- Network rate, per protocol year
- 3.556%
- Per calendar year
- 3.37%
- Staked share of supply
- 67.3%
Solana pays its stakers in new SOL. Every epoch the protocol creates a slice of new coins and hands it to stake accounts in proportion to their stake and how well their validator voted. Nobody votes on the amount each epoch: a schedule fixed at launch sets it.
The schedule, read from the network
The schedule’s parameters are stored in the chain and returned by the getInflationGovernor RPC call. On November 16, 2026 it returned:
| Parameter | Value | Meaning |
|---|---|---|
initial |
0.08 | 8% a year when inflation was switched on |
taper |
0.15 | each year the rate falls by 15% of itself |
terminal |
0.015 | the rate never falls below 1.5% a year |
foundation |
0 | no share goes to the Solana Foundation |
Inflation was switched on at epoch 150, on 10 February 2021 (SIMD-0228’s account of it). So the rate is 8% × 0.85 raised to the number of protocol years since then, until it hits 1.5%.
Where it stands: 3.556% per protocol year at epoch 1,087 (getInflationRate, 2026-11-16 00:42 UTC), which is about 3.37% of supply per calendar year (computed below). From the protocol rate, reaching the 1.5% floor takes ln(1.5 / 3.56%) ÷ ln(0.85) = about 5.3 protocol years (computed).
Protocol years and calendar years
The protocol does not read a clock to decide what year it is; it counts slots. A protocol year is a fixed number of slots: 78,892,315 at the old 400 ms target, and twice that, 157,784,630, since slots were cut to 200 ms (SIMD-0525 doubled it so the yearly figure would be unchanged at exactly 200 ms). getInflationRate reports a rate per protocol year, and each epoch issues that rate × 432,000 ÷ 157,784,630 of total supply.
Real slots run slower than the target (215 ms on November 16, 2026), so a protocol year takes longer than a calendar year: a calendar year holds about 340 epochs, not about 365. Compounding the per-epoch issuance over the epochs that really fit in a calendar year gives about 3.37% of supply per calendar year, against the 3.556% the network reports. That is the figure that describes how fast supply grows, and the one this site uses for every “per year” number; our methodology has the formula. The same slot clock explains the schedule: five and three quarter calendar years after launch, the rate is where an exact calendar schedule would have put it in early 2026.
Who gets the new SOL, and what it really pays
All of it goes to stakers, through their validators. Each epoch’s issuance is spread over the 67.3% of supply that is staked, so each epoch paid staked SOL about 0.0145% before the validator’s commission. Compounded over the 340 epochs in a calendar year, that is 5.04% a year on staked SOL before commission, and 4.79% at a 5% commission (computed from reads on November 16, 2026; tips not included).
That is not the same as getting richer by 4.79%. Everyone’s SOL is diluted by the new coins. A staker’s share of all SOL grew by about +1.37% a year; an unstaked holder’s share shrank by about -3.26%. Our real yield page shows the three numbers side by side and why an “APY” without the dilution overstates what staking does for you.
Two small things work the other way. Half of every 5,000-lamport base fee is burned, and since Alpenglow each validator has 0.8 SOL burned from its vote account every epoch as an admission ticket. Together these offset only a small part of issuance (the where fees go page measures it).
What was proposed, and what happened
| When | Proposal | What it would do | Outcome |
|---|---|---|---|
| Jan to Mar 2025 | SIMD-0228, “market-based emission” | Cut issuance when more SOL is staked, with a curve based on the square root of the staked share | Vote failed to reach quorum (as SIMD-0550’s text records); pull request closed unmerged on 2 May 2025 |
| Nov 2025 to Jan 2026 | SIMD-0411 | Double the yearly decline from 15% to 30% | Closed on 31 January 2026 without a vote |
| Jun 2026 onward | SIMD-0550, “Double Disinflation” | The same 30% decline, starting from whatever the rate is when it activates, with the 1.5% floor kept | Document merged 23 July 2026, status “Review”. A governance vote, SGP-0002, is reported to have passed on 28 August 2026 with 67% of participating stake and 60.7% turnout (crypto.news, secondary). Not active on-chain. |
Where Double Disinflation actually stands
The vote result above comes from a news report; the governance dashboard renders in a browser and we did not read it directly. What we did check on November 16, 2026:
- The feature that would switch it on,
double_disinflation_rate(idddibiks7RHZR69QXYMCmk4EZNt6WckspTqEEWv6cS38in Agave’s feature list), does not exist as an account on mainnet, so it is not even queued. getInflationGovernorstill returns a taper of 0.15. SIMD-0550 says the taper would read 0.30 after activation.- SIMD-0550 requires another change, SIMD-0607 (integer maths for inflation), to activate first. SIMD-0607 is a draft.
If it does activate, SIMD-0550’s authors estimate the floor would arrive in about 2.8 years instead of about 5.7 (their figures from June 2026), removing roughly 18.9 million SOL from issuance over six years. The rate would not jump on the day: it would start from where it is and fall twice as fast.
What this means for you
If you hold SOL and do not stake it, your share of the network shrinks by about -3.26% a year at today’s rate; staking, natively or through a liquid staking token, is how holders avoid that. The rate is not a promise: it falls every epoch on schedule, and it would fall faster if Double Disinflation is ever activated. We re-read it every two hours and keep the history on the supply page.
Questions people ask
What is Solana's inflation rate right now?
The network reported 3.556% per protocol year on November 16, 2026 (getInflationRate). Per calendar year, at the measured epoch length, that is about 3.37% of supply. It falls a little every epoch; our supply page keeps the daily history.
Does Solana have a max supply?
No. Total supply was 637.6 million SOL on November 16, 2026 and grows every epoch. The schedule caps the rate, not the amount: once it reaches 1.5% a year it stays there, minus whatever fees and admission tickets burn.
What happened to SIMD-0228?
SIMD-0228 proposed cutting issuance when more SOL is staked, using a curve based on the square root of the staked share. Its vote in March 2025 failed to reach quorum (as SIMD-0550's own text records), and the proposal was closed unmerged on 2 May 2025. The fixed schedule stayed.
Is Double Disinflation active?
No. A news report says the governance vote SGP-0002 passed on 28 August 2026 with 67% of participating stake, but the feature that would apply it, double_disinflation_rate, did not exist on mainnet when we checked on November 16, 2026, and SIMD-0550 says another change (SIMD-0607) must activate first.
Did faster blocks change inflation?
Not on paper, slightly in practice. SIMD-0525, which cut slots from 400 ms to 200 ms, also doubled the protocol's slots-per-year figure so issuance per year would stay the same if slots ran at exactly 200 ms. They run at about 215 ms, so a calendar year holds about 340 epochs instead of about 365, and issues about 3.37% of supply rather than the 3.556% the network reports.
Sources
- Solana RPC: getInflationGovernor and getInflationRate (read 2026-11-10)
- SIMD-0228: Market-Based Emission Mechanism (pull request, closed unmerged 2 May 2025)
- SIMD-0550: Double Disinflation Rate (status Review)
- SIMD-0411: Double Disinflation Rate (earlier version, closed 31 Jan 2026)
- SIMD-0525: Reduce Slot Times (slots_per_year doubled to 157,784,629.968)
- Agave feature set: double_disinflation_rate feature id (not present on mainnet, 2026-11-10)
- crypto.news: Solana inflation cut clears vote with 67% support (secondary)