Total supply is every token of a project that exists right now, minus any that have been verifiably burned. It includes tokens nobody can sell yet: pre-mined allocations held back deliberately and balances locked inside vesting contracts. That makes it the count of the obligation rather than the count of the market, which is why market cap uses circulating supply instead and why the gap between the two figures is the part worth reading.
Tokens that arrive at a vesting cliff do not appear from nowhere. They were in total supply the whole time. The cliff only moves them across the line into circulation, which is where price impact happens.
Scroll to see the full diagram
Total supply counts tokens that exist and are not for sale
CoinMarketCap defines total supply as the total amount of coins in existence right now, minus any coins that have been verifiably burned.1 The definition turns on existence, not availability. A token minted at genesis and locked in a four year vesting contract is fully counted in total supply from day one, even though no holder can move it.
The same source is explicit about what the bucket contains: pre-mined coins intentionally kept out of circulation, and tokens locked in smart contracts until a particular purpose has been achieved, with ICO vesting stages given as the example.1 Both are ordinary features of a well structured launch rather than warning signs.
So total supply answers a narrow question. How many units has this project committed to bringing into existence and not yet destroyed. Read next to circulating supply it becomes the schedule of everything still to come.
Only verifiable burns come out
The word doing the work in the definition is verifiably. A burn counts against total supply when anyone can check it onchain: tokens sent to an address with no known private key, or destroyed through a burn function that reduces the contract's own supply variable. A team announcement that tokens have been retired does not reduce total supply until the transaction exists and can be read.
The academic treatment of this is more interesting than the mechanics. An NBER working paper by Cong, Li and Wang models issuance and burning as endogenous levers rather than housekeeping, with tokens optimally issued to reward platform owners when the productivity-normalized token supply is low, and burnt to boost franchise value when that supply is high.3 Whatever you make of the model, the framing is the right one: supply changes are a financing decision, not an accounting detail.
Two practical consequences follow. If you intend to burn, write the mechanism into the contract so the reduction is checkable rather than announced. And expect providers to keep any burn out of their figures until it settles, which means your reported total supply will lag your intent.
BNB shows a total supply that moves downward
Read on 3 August 2026, CoinGecko's record for BNB carried a total supply of 133,164,946.5 tokens against a max supply field of 200,000,000, a gap of nearly 67 million tokens.2 The total is a fractional number, which is itself a reminder that these are contract balances rather than round design figures.
The direction of travel matters more than the number. Most tokens have a total supply that rises as emissions mint new units. A token running programmatic burns has one that falls, and providers then disagree about whether the ceiling above it falls too, which is the argument covered on the max supply page.
For anyone building a model this is the trap. Do not treat total supply as a constant you can hardcode. Pull it as a series, plot it, and note which direction it moves and why, because a supply schedule that only exists as a paragraph in a whitepaper will not match what the contracts have actually done.
The cliff illusion
The most common misreading of these figures is treating a vesting cliff as new supply. It is not. Every token released at a cliff was already inside total supply, already counted in FDV, and already visible on the schedule the project published. Nothing was created on the day it unlocked.
What changed is which side of the circulating line those tokens sit on. That is the only thing that matters for price, because market cap, order book depth and the ability to sell all operate on the circulating side. The token supply is unchanged; the tradeable supply is not.
This is why we push founders to model the float curve rather than the supply curve. The supply curve is usually flat or gently rising and looks reassuring. The float curve has the steps in it, and the steps are where the treasury either has a plan or does not.
What belongs in the tokenomics document
Three lines that most documents leave out. The total supply at genesis and whether it can increase, stated as a property of the contract rather than an intention. The burn mechanism, if any, with the function and the address, so the reduction is auditable rather than reported. And the mapping from total supply to circulating supply over time, month by month, with the contract addresses that enforce each release.
Write those three and most of the questions a data provider, an exchange listing team or a diligence process will ask are already answered. Leave them out and each of those parties will reconstruct an answer from what they can observe, and their reconstruction becomes the number the market uses.
The underlying point is the same one that governs every supply metric. The token is infrastructure for a business, and every unit in total supply is a claim on whatever that business ends up producing. Issuing more of them is cheap. Growing into them is not.
Common questions
What is total supply in crypto?
Total supply is every token of a project that currently exists, minus any verifiably burned. CoinMarketCap defines it as the total amount of coins in existence right now less coins that have been verifiably burned, and notes it includes pre-mined coins held out of circulation and tokens locked in smart contracts such as during ICO vesting stages.1 It counts existence, not availability.
What is the difference between total supply and circulating supply?
Total supply includes locked and inaccessible tokens; circulating supply is the subset judged available in the market. Market cap uses circulating supply because prices are driven by tokens that can actually be used rather than those sitting inaccessible in a contract.1 The gap between the two is supply already committed and scheduled to reach the market on a published timetable.
Can total supply decrease?
Yes, through verifiable burns. Tokens sent to an address with no known private key or destroyed through a contract burn function reduce total supply, while an announcement alone does not. Reading BNB on 3 August 2026, CoinGecko reported a total supply of 133,164,946.5 against a max supply field of 200,000,000, so the reported total sat well below the reported ceiling.2
Does total supply include locked tokens?
Yes. That is the main thing separating it from circulating supply. Tokens sitting in a vesting contract, a treasury multisig or a pre-mined allocation held deliberately out of the market are all inside total supply from the moment they are minted. They only move into circulating supply when the lock releases, which is why a cliff changes tradeable supply without changing total supply at all.
See Token Allocation and Vesting for how this applies in practice.
Sources
- Total Supply, CoinMarketCap Academy glossary
CoinMarketCap, 2026
Definition of total supply, its treatment of pre-mined and contract-locked tokens, and why market cap uses circulating supply instead. - CoinGecko public API v3, coin record for BNB
CoinGecko, 2026
Total supply of 133,164,946.5 against a max supply field of 200,000,000, read on 2026-08-03. Point-in-time values. - Token-Based Platform Finance, NBER Working Paper 27810
National Bureau of Economic Research, 2021
Models token issuance and burning as endogenous supply-management levers tied to platform productivity. NBER working paper, September 2020 revised May 2021.
Last reviewed 2026-08
Know the terms but not sure how they apply to your project? That is what an engagement is for. We design, document, and stress-test the whole token economy inside the Tokenomics Data Room.
100+ projects advised. Complete tokenomics in 4 to 6 weeks.