Burn-Mint Equilibrium: What Actually Changes for AKT Holders

What is happening

On 23 March 2026, Akash Network activated AEP-76: Burn-Mint EquilibriumBurn-Mint EquilibriumA tokenomics model where network fees burn tokens while new tokens are minted and paid to suppliers. The system tries to balance burns and mints so circulating supply stays roughly stable when usage scales.Like a business that spends a dollar of revenue for every dollar of wages it pays. Money flows in and out at the same rate, so the total cash in the company stays flat. The rate of flow tells you how big the business is.Read more →, as part of its Mainnet 17 upgrade (Proposal 318, v2.0.0). The on-chain governanceDAODecentralised Autonomous Organisation. A way to coordinate decisions and manage a treasury using token-weighted voting instead of a traditional company structure. Token holders propose and vote on changes directly.Like a shareholder-run company where every shareholder can vote on every decision, the votes are public, and the company can't do anything the shareholders don't approve. The coordination is messier than a normal company but nobody has unilateral control.Read more → vote cleared with 99.7% in favour, and mainnet moved to a burn-and-mint model with WASM smart contractSmart ContractA program stored on a blockchain that runs automatically when its conditions are met. Smart contracts are how blockchains do anything beyond just transferring tokens — DeFi, NFTs, DAOs, and DeAI infrastructure all run on smart contracts.Like a vending machine. You put in the right input and it produces the expected output, no human operator required. The rules are fixed in the machine itself, anyone can use it, and nobody can stop a transaction in the middle.Read more → support.

This was the single most significant change to AKT tokenomics since the network launched in 2020. It was designed to transform AKT from a straightforwardly inflationary stakingStakingLocking up a cryptocurrency to help secure a blockchain network, usually in exchange for rewards. The locked tokens act as a security deposit that can be taken away if the staker misbehaves.Like putting down a large rental deposit for an apartment. You get the money back if you behave, you earn interest while it's locked, and the landlord takes it if you trash the place.Read more → tokenTokenA digital unit of value or access rights tracked on a blockchain. Tokens can represent ownership in a project, a right to use a service, a share of future revenue, or simply a tradable asset with no underlying claim.Like a physical poker chip a casino issues. The chip itself has no value. What makes it worth something is what it lets you do at the casino, what the casino has promised, and how much other people will pay you for it.Read more → into a burn-and-mint asset where compute spending creates direct demand for the token. Whether it has done that at current usage is the question this piece works through.

I hold AKT. I also have indirect exposure through Venice (which uses Akash GPUs) and Morpheus (which routes compute through Akash). This article explains what BME actually does and what it changed for holders like me.

How AKT worked before BME

Before BME, the tokenomics were simple and not particularly attractive:

• 8% capped annual inflationInflationThe annual rate at which new tokens are created and added to the circulating supply. Most networks use inflation to pay validators, stakers, and infrastructure providers from freshly minted tokens rather than real revenue.Like a landlord who raises the rent every year. If your salary goes up at the same rate, you break even. If it doesn't, you get poorer without noticing, because the number on your payslip hasn't changed but the ground under it has shifted.Read more →. New AKT is minted every blockBlockA batch of transactions added to a blockchain at a set interval. Each block cryptographically links to the previous one, creating an append-only chain that can't be rewritten without redoing all the work since.Like a page in a ledger. Every page has a fixed number of entries, every page references the previous page, and once a page is filled and signed off it can't be edited without visibly invalidating every page that came after. The chain is just a very long series of these sealed pages.Read more → as staking rewards, capped at 8% under Proposal 283. No burnBurnPermanently removing tokens from circulation by sending them to an address that no one controls. Burns reduce total supply, which (all else equal) makes each remaining token worth more of the network's value.Like a company buying back its own shares and shredding them. The company's total value stays the same, but each remaining share now represents a slightly bigger slice of that value.Read more → mechanism offset it.
• Take-rate fees. The protocol charged 4% on AKT compute payments and 20% on USDC payments. These fees went to the community pool. The amounts were negligible at the network’s revenue.
• Staking yield around zero after inflation. Nominal staking yield sat close to the inflation rate, so after the 50% community pool tax, real yield was roughly nothing. You stake to avoid dilution, not to earn.
• No link between usage and token value. Someone deploying compute on Akash created no direct buy pressure or burn on AKT. The token and the product were economically disconnected.

That last point was the fundamental problem. Akash has verifiable compute revenue, more than most DeAIDeAIDecentralised AI. An umbrella term for blockchain-based projects that build AI infrastructure (compute, data, inference, models, agents) without a single central provider controlling the system.Like the difference between streaming a movie from Netflix and sharing it via BitTorrent. Netflix is fast and polished but one company controls what you can watch and what you pay. BitTorrent is messier but no single operator can shut you out.Read more → projects can claim. But that revenue didn’t flow through the token in a way that benefits holders. BME was designed to change this.

How BME works

The mechanism has two phases: mint and settle.

Mint phase (tenant top-up)

When a tenant funds compute on Akash:

1. They provide AKT (directly or via credit card conversion)
2. The protocol burns the AKT at the current oracle price
3. New ACT (Akash Compute Token) is minted: ACT minted = AKT burned × Oracle Price
4. ACT is pegged at 1 ACT ≈ $1 of compute credit

ACT is soulbound: non-transferable, no expiration, fully refundable back to AKT at the current oracle price. Think of it as a compute escrow denominated in dollars but collateralised by AKT.

Settlement phase (provider payout)

When a provider completes work and gets paid:

1. The outstanding ACT is burned
2. AKT is reminted to the provider based on the settlement TWAP (time-weighted average price)
3. Net supply change depends on price movement: ΔSupply(AKT) = 1/P_mint − 1/P_settle

This is the key formula. If AKT appreciates between when the tenant tops up and when the provider gets paid, fewer AKT are reminted than were burned. Net burn. If AKT depreciates, more are reminted. Net inflation.

Worked example

A tenant funds $1,000 of compute when AKT is $1.14:

• 877.19 AKT burned, 1,000 ACT minted

Provider completes the work. Settlement happens when AKT is $1.50:

• 666.67 AKT reminted to the provider
• Net burn: 210.53 AKT (~$240 worth permanently removed from supply)

In a rising market, BME is deflationary. In a falling market, it is inflationary. In a flat market, it is roughly neutral. The mechanism amplifies the directional trend.

The safety infrastructure

This is more than a naive burn. Akash built a BME vault architecture with concrete guardrails:

• BME vault holds AKT from mint operations. The vault is depleted before new minting occurs, minimising unnecessary inflation.
• Collateral ratio monitoring: CR = (Vault AKT × Price) / Outstanding ACT. If the vault’s AKT drops in value relative to outstanding compute credits, circuit breakers trigger.
• Circuit breakers: Warning at CR 0.95. Halt new ACT mints at CR 0.90. This prevents the system from over-minting during sharp price drops.
• Dual-feed oracle: 30-minute TWAP from Osmosis plus Pyth/Chainlink, with outlier rejection at >1.5% divergence. No single oracle can manipulate the price feed.

The vault and circuit breaker design is more conservative than most DeFiDeFiDecentralised Finance. Financial services like lending, trading, and yield farming built on smart contracts instead of traditional banks or brokerages. DeFi protocols are usually permissionless and global.Like a vending machine that can give you a loan, swap your currencies, or invest your savings. Nobody is behind the counter, the rules are written into the machine itself, and anyone with money in the right format can use it.Read more → burn mechanisms. It’s engineered to prevent catastrophic scenarios where a price crash forces massive AKT inflation.

What changes for holders

For stakers

Your staking mechanics don’t change. Inflation parametersParametersThe internal numbers (weights and biases) inside a neural network that get adjusted during training. A 70-billion-parameter model has 70 billion adjustable internal numbers encoding everything it has learned.Like the synapses in a human brain. Each parameter is a tiny dial that gets nudged a little during training. With enough dials, the network can represent surprisingly complex patterns. The total parameter count is roughly how much "brain" the model has.Read more → remain governed separately. What BME changed is the relationship between usage and dilution.

Here is the honest part. The pre-launch projection was that BME would cut effective inflation by roughly a third at then-current usage. That hasn’t happened.

Messari reports 53,520 AKT burned in the first nine days after activation, roughly 5,950 a day, which offsets only a small fraction of issuance. Annualised inflation actually ran at 8.94% in Q1 2026, above its 8% cap and up from 8.12% the prior quarter, not down toward the projected 5%. Lease revenue fell over the same period. The dilution relief BME was supposed to deliver has been marginal so far.

What did change is structural, and that still matters. Before BME, more Akash usage didn’t help stakers at all. After BME, every dollar of compute revenue reduces the dilution stakers face. The lever now exists. The volume flowing through it has simply been too small to move the headline inflation number.

For providers

Providers are paid in AKT at settlement TWAP, with zero take-rate. Better than the old system in two ways:

1. No more take-rate. The old 4% AKT / 20% USDC fees are eliminated. Replaced by a 25 basis point mint-side spread, a fraction of the old cost.
2. AKT payment creates alignment. Providers now receive AKT, making them token holders with skin in the network’s success. Previously, providers accepting USDC had no direct token exposure.

For traders

BME creates a direct link between compute demand and AKT buy pressure. Every tenant deployment triggers an AKT market buy to fund the burn. The catch is scale: that buy pressure is only as large as the compute revenue behind it, and at current revenue it remains modest.

The 25bps spread on the mint side means the bulk of every dollar spent burns AKT directly. The remainder goes to the BME vault buffer.

What does not change

• Staking rewards and inflation schedule. Still governed by separate parameters.
• 21-day unbonding period. Unchanged.
• Governance mechanics. Unchanged.
• Provider onboarding. Still permissionless.

How this compares to Render’s BME

Both Akash and Render use burn-mint equilibrium, but the implementations are fundamentally different. I covered this in RENDER vs AKT vs IO: The Revenue Question, but the BME specifics are worth comparing directly.

The key conceptual difference: Render decouples payment from reward. When a creator burns RENDER, providers are paid from a separate fixed emission schedule proportional to work done. The burn and the provider payment are independent flows.

Akash couples them. The AKT burned by tenants is effectively the same AKT reminted to providers. Supply change is a direct function of price movement between mint and settlement. This is more like a collateralised stablecoin mechanism than a traditional burn.

Which is better? Neither, necessarily. Render’s modelModelA trained neural network that takes inputs (text, images, audio) and produces outputs (more text, classifications, generated content). In DeAI the model is the thing that actually does the work.Like a very experienced apprentice who has spent years watching thousands of masters make furniture. They can't explain how they know when a joint is right, but they can make a chair that looks and functions like a Chippendale. The training is invisible. The output is what matters.Read more → is simpler and has a hard supply cap. If usage grows enough, deflation is inevitable. Akash’s model is more dynamic. Burn intensity depends on price direction, which introduces more complexity but also means a rising AKT price actively accelerates the deflationary effect. In a bull market, Akash’s BME burns harder. In a bear market, it inflates harder.

Render’s 8x gap between emissionsEmissionsNew tokens created and distributed by a blockchain protocol over time as rewards to validators, stakers, or miners. Emissions fund network security and participation at the cost of diluting existing holders.Like a company that pays employees partly in newly printed shares. Every year the total number of shares goes up, which means existing shareholders own a slightly smaller slice of the same company unless the company grows faster than the printing.Read more → and burns is a structural challenge. Akash’s gap is narrower but the target is moving because there is no supply cap.

What I’m watching

Five metrics that matter now BME is live:

1. Monthly AKT burned vs reminted. The net burn (or net mint) is the headline number. Positive net burn means the system is working as designed at current price trends.

2. BME vault collateral ratio. Should stay well above 1.0. If it approaches 0.95, the market is testing the mechanism. Below 0.90 and circuit breakers halt new ACT mints, which would be a stress scenario.

3. Compute revenue growth. This is the lever that makes BME matter. At current revenue, burns offset only a fraction of issuance, which is why the projected inflation cut hasn’t shown up. The mechanism only bites if compute spend scales by roughly an order of magnitude.

4. Provider count and utilisation. BME should improve provider economics (no take-rate, AKT alignment). If provider count starts growing after its 2025 decline, BME is having the intended effect on supply-side participation.

5. ACT outstanding. Large ACT balances mean significant AKT is locked in the escrow mechanism, reducing effective circulating supplyCirculating SupplyThe number of tokens currently in circulation and tradeable on the open market. Differs from total supply (which includes locked or unvested tokens) and max supply (the upper limit, if there is one).Like the number of cars on the road today versus the number ever produced. Some are in showrooms, some in junkyards, some still at the factory. Only the ones on the road count toward what people are actually driving.Read more →. This is an under-appreciated side effect. ACT acts as a temporary AKT sink.

The honest assessment

BME is a clear improvement to AKT tokenomics. It takes a token that had no meaningful connection between usage and value and creates a direct, on-chain, verifiable link. Every dollar of compute spent now burns AKT. That link is on-chain and auditable.

But let’s not pretend it solves everything. The pre-launch projection that BME would meaningfully cut inflation hasn’t held at current usage: annualised inflation ran at 8.94% in Q1 2026, and realised burns offset only a fraction of issuance. The token is still inflationary. Still no supply cap. The mechanism needs significantly higher revenue (an order of magnitude, realistically) before net deflation becomes plausible.

The comparison to Render is instructive. Render has the same fundamental challenge in reverse: a hard supply cap and a proven burn mechanism, but burns that lag emissions by 8x. Both networks are betting that compute demand will grow fast enough to flip their respective models deflationary. Neither is there yet.

What BME does accomplish is alignment. Before, Akash had a product (compute marketplace) and a token (AKT) that were economically disconnected. After BME, the product drives the token. That matters for the investment thesis even before the numbers are large enough for net deflation.

BME is now live, and we’ve updated our Akash project review accordingly: the returns score moved from 62 to 68, primarily on value accrual and supply dynamics. The mechanism works and it burns. The open question is whether demand scales enough to make the burn matter.