Originally published at mrnasdog.com/research/inj/inflation by MrNasdog.
Injective's staking inflation created 1.03M INJ over the last 90 days while the Community BuyBack destroyed 135.4K INJ across four settlements — a real burn, and roughly one eighth of the mint it offsets. The MrNasdog Pressure Framework reads INJ at +0.89% net for the trailing quarter and +1.02% forward, against a supply-monitor reading of −0.10%: a gap of 0.99 percentage points that comes from the counted float, not from Injective. INJ is mildly inflationary on an uncapped staking mint, and the burn narrows the gap without closing it.
The verdict, in one paragraph
Over the 90 days to Sep 4 2026 the framework reads Injective at +0.89% net: 1.03M INJ of new staking issuance against 135.4K INJ destroyed, on a counted base of 100.0M INJ. The supply monitor reads the same window at −0.10%, a gap of 0.99 percentage points, which is over tolerance and ships with a data-conflict flag. The reason is specific and checkable: the counted circulating figure for INJ is pinned at exactly 100,000,000 and has not moved once across the monitor's entire record, while the chain itself read 122,719,892 INJ at block 181,571,003 and adds 0.0856 INJ in every block. A frozen denominator cannot register issuance, so the monitor reports noise around zero where the chain reports growth. Forward, the framework reads +1.02%, because blocks have got faster since the July and September upgrades while the buyback rounds have got smaller. Injective is structurally inflationary on the active float, with a genuine but under-sized burn.
Sell pressure: where new INJ comes from
Sell #1 — protocol inflation — is the whole sell side at 1.03M INJ, and it is Injective's staking-rewards stream. The rate is not fixed: Injective targets a 60% bonded ratio and moves the annual rate between a 2.2% floor and a 4.4% ceiling depending on how much INJ is staked. Only 58,863,505 INJ of the chain's 122,719,892 is bonded — 47.97%, well under the target — so the rate has been pinned at the ceiling all window. The headline number flatters, though. Injective pays its staking reward per block against a schedule that assumes a 0.5-second block, and this build counted the blocks rather than trusting the target: 12,055,026 of them between heights 169,515,977 and 181,571,003, which is a 0.645-second block. Realised issuance is therefore 3.41% a year, not 4.4%. The per-block figure itself was measured, not modelled: two reads 101 blocks apart give 0.085611348 INJ per block against a schedule of 0.085611293.
Sell #2 — vesting unlocks — is zero, and permanently so. Injective's genesis release schedule ran out in January 2024; the original 100 million INJ allocated at launch is fully distributed, and there is no cliff, no linear stream and no calendar entry left that can land in any future window. Sell #3 — foundation and unscheduled unlocks — is also zero. The only team-controlled pot on the Injective chain is the DAO community pool, which held 17,468.60 INJ when this build read it. That pool is filled by a 5% skim on the staking mint already counted in Sell #1, and it is emptied by public vote — three liquidity-provider payouts passed inside the window — so it recycles supply the ledger has already booked rather than adding new supply. Sell #4 — long-term locked or bankruptcy — is zero: Injective has no bankruptcy estate and no trustee distribution, and its 21-day unstaking delay is a queue rather than a lock, sitting inside the counted float either way.
Buy pressure: where new INJ goes
Buy #1 — programmatic buyback — is the entire buy side at 135.4K INJ, and it is the most interesting row on the page. Injective routes 60% of the trading fees its applications generate into a basket, then auctions that basket to a bidder who pays in INJ, and the winning INJ bid is destroyed. Since November 2025 the auction has run on a 28-day cycle as the Community BuyBack: participants commit INJ, receive a share of ecosystem revenue in return, and every committed token is burned at settlement. Four rounds landed inside this window — 39,300 INJ on Jun 10 2026, 43,500 on Jul 8 2026, 27,400 on Aug 5 2026 and 25,200 on Sep 2 2026 — summing to 135,399.41 INJ.
Injective has no burn address, which is exactly why the burn has to be proved rather than quoted. Nothing is sent anywhere; the auction module destroys the coins in place, so the only surface that can show it is total supply itself. This build located the settlement block by bisection and watched it happen: at block 181,266,884, timestamped Sep 2 2026, Injective's total supply fell by 25,199.622253 INJ net of that block's own mint, against a winning bid of 25,199.622251 INJ read independently from both the auction module and the buyback contract. Three surfaces, one number, and they agree to six decimal places. The identity is the proof: had the committed INJ merely moved into another wallet, supply would have risen by the full mint. Across the surrounding 540,000 blocks the chain minted 46,230 INJ and supply rose only 6,462.
Buy #2 — protocol fee burn — is zero, and evidenced rather than assumed. Injective has no base-fee mechanism at all, and the account that collects gas fees returns an empty balance at every moment because it is swept every block. Independently, across a quiet 45,000-block stretch with no auction in it, supply tracked the mint schedule to within 0.38 INJ — there is no continuous burn hiding underneath. The trading-fee share that does fund destruction is the basket Buy #1 bids for, so it is counted once. Buy #3 — foundation buy — is zero: the only publicly-traded INJ treasury bought 560,647 INJ in February 2026 and has not bought since, and an unexercised ability to buy is not a flow. Buy #4 — new long-term lock — is zero: bonded INJ grew, but staking sits inside the counted float, so bonding removes nothing from it.
Foundation and overhang
Injective's team-controlled overhang is unusually small for a chain of its size, and it is enumerated here in full. First, the DAO community pool at 17,468.60 INJ — on-chain, readable at any time, refreshed on every rebuild, and spent only by public vote. Second, the buyback destination, which is not an overhang at all: bought-back INJ is destroyed rather than accumulated, and there is no wallet behind it to watch. Third, the corporate treasury: a listed company holding INJ bought 560,647 INJ in February 2026 and states it stakes what it holds, which is tracked through its own regulated filings rather than through a published address. There is no foundation reserve with a release schedule, no unscheduled allocation and no bankruptcy residual, because the genesis distribution finished in January 2024.
There is a fourth item that behaves like an overhang without being one. INJ exists on two ledgers: a fixed 100,000,000 token on Ethereum, of which 94,767,530 sits locked in Injective's bridge, and the 122,719,892 native INJ on Injective itself. Adding the unlocked Ethereum remainder to the native supply gives a real global figure of 127,952,363 INJ. The bridge mints and destroys native INJ as coins cross, which is why the raw chain supply moves in both directions beyond the mint schedule, and why this build measured the mint and the burn separately rather than differencing supply across the quarter. If any of these balances falls between refreshes, the outflow enters Sell #3 at the next refresh.
How INJ compares to other uncapped proof-of-stake L1s
Injective sits in the same structural family as every uncapped Cosmos-style chain that targets a bonded ratio: the rate is a dial, not a constant, and it rises when staking participation falls. That is the opposite of a halving chain, where issuance is a fixed decreasing schedule nobody can move, and it is also the opposite of a hard-capped token, where the sell side is a distribution question rather than an issuance question. What separates Injective from its nearest peers is the buy side. Most bonded-ratio chains have none at all — new coins are minted, paid to stakers and that is the end of it. Injective takes a fixed share of application revenue and destroys INJ with it, which puts it closer in shape to an exchange token running a revenue-funded buy-and-burn than to a plain staking L1.
The catch is proportion. A revenue-funded burn only turns a chain deflationary when revenue is large relative to issuance, and on Injective it is not: 135.4K INJ destroyed against 1.03M INJ minted is a coverage ratio near 13%. Exchange tokens that genuinely shrink their supply typically run zero issuance, so every burned coin is a net subtraction; Injective is subtracting from a growing base. It also differs from base-fee-burn chains, where the burn scales with raw transaction volume — Injective's burn scales with application trading fees, a narrower and more cyclical input, and the last two settlements have stepped down from 43,500 to 27,400 to 25,200 INJ.
One more comparison matters for anyone reading INJ against its peers on a supply chart. Bonded-ratio chains are usually measured against a float that tracks the chain, so the published inflation and the observed float move together. Injective's do not: the counted circulating figure is the fixed Ethereum token supply, so a chart of INJ's float shows a flat line through four years of real issuance. Any comparison drawn from that chart understates Injective's dilution by roughly 28% of supply.
What to watch in the next 90 days
First, the buyback settlements on Sep 30 2026, Oct 28 2026 and Nov 25 2026 — three rounds fall inside the forward window and their sizes have been falling, so a round back above 40,000 INJ would visibly improve the net. Second, the bonded ratio: staking sits at 47.97% against a 60% target, and every point of extra bonding pulls the rate down off its 4.4% ceiling toward the 2.2% floor. Third, block times, which drive issuance directly on Injective — the chain averaged 0.645 seconds over the trailing quarter and 0.609 seconds since Aug 6 2026, and faster blocks mean more INJ per day, not less. Fourth, the staked-INJ exchange-traded fund filed on Aug 24 2026: if approved, it would lock INJ in custody and open a Buy #4 row that is currently empty. Fifth, governance — this build walked every proposal from Jun 1 2026 to Sep 4 2026 and found no change to the issuance parameters, so any new proposal touching them would re-base the whole page.
Summary
Injective mints INJ continuously and destroys some of it on a 28-day cycle, and the mint is the larger of the two by roughly eight to one. Over the 90 days to Sep 4 2026 staking rewards created 1.03M INJ while four buyback settlements destroyed 135.4K INJ, leaving the MrNasdog Pressure Framework at +0.89% net, rising to +1.02% forward as blocks speed up and rounds shrink. The burn is genuine — total supply falls by the exact bid at the settlement block, with no address to accumulate behind it — but it is not large enough to make INJ deflationary, and nothing in the protocol caps supply. The single biggest thing to know is that Injective has no supply cap and its published float does not track its chain: the counted figure is 100,000,000, while real global INJ supply is 127,952,363 and rising every block.
MrNasdog Pressure Framework analysis of INJ, Metric 1 — Inflation. Data + explanation only. Not financial advice. Updated Sep 5 2026.
Top comments (0)