What is Starknet? Reasons for the Supply Cap of 10 Billion in Published Information

Starknet is a Layer 2 network that processes transactions outside of Ethereum, proving their correctness and sending verification proofs back to Ethereum. STRK is the native token used for transaction fee payments, staking, and voting on protocol changes.

Major price aggregators list the total supply of STRK as 10,000,000,000, with the maximum supply also stated as the same figure. However, on September 9, 2026, when reading the Starknet Ethereum contract at block 25,940,909, 10,162,500,000 STRK was returned. This is 162,500,000 STRK more than the stated amount. Additionally, the official documentation from the issuer explains that there is no cap on the maximum supply.

According to Starknet's official documentation, it is described as a decentralized, permissionless Layer 2 validity rollup that scales Ethereum using zero-knowledge proofs. It aggregates thousands of transactions outside of Ethereum and finalizes them on Ethereum after verification. If you have read about how Ethereum's Layer 2 solutions work, Starknet is one of them.

Starknet has two design features that differentiate it from other networks. In Starknet, all accounts are smart contracts themselves. Therefore, the rules for transaction approval are not fixed by the protocol but are written in software. This allows for the use of multiple signers, time-limited session keys, and passkeys in wallets without changing the network.

Another feature is the fee token. Previously, Starknet fees could be paid in Ether. It later changed to allow payment in either Ether or STRK. Since the release of version 0.14.0 on September 1, 2025, the official documentation states that transaction fees can only be paid with STRK. This is akin to a toll booth that accepts only one type of currency. To use the network, one must hold STRK for fees.

Reasons Why the Supply of 10 Billion is No Longer Accurate

Almost all STRK listing pages display both the total supply and maximum supply as 10 billion. However, neither figure can be verified on the contract.

Here’s how to reproduce the findings. I called the total supply function of the Ethereum contract at block 25,940,909 with a timestamp of September 9, 2026, 16:08 UTC. I also checked the Starknet contract at block 14,613,788 at 16:06 UTC on the same day. Both addresses are listed in the table above. Three independent Ethereum nodes and two Starknet nodes returned the same values.

The Ethereum contract returned 10,162,500,000 STRK, which exceeds 10 billion by 1.625%. The Starknet contract returned 3,805,293,274.848997 STRK. This is a separate figure indicating different subjects, and the discrepancy is expected. The Starknet figure counts only the amount that has been bridged and moved, not the total tokens. Additionally, aggregators list a third figure of 7,181,521,515 STRK as the circulating supply. None of these three figures amount to 10 billion.

The part regarding the supply cap is not just outdated but clearly incorrect. According to Starknet's documentation, 10 billion tokens were created in May 2022 and minted on-chain on November 30, 2022. There are explanations omitted from the listing pages. The protocol mints new tokens for staking rewards, so the total supply increases over time. The curve determining the minting rate is also published. The minting rate is calculated by multiplying the value obtained by dividing 4% by 10 with the square root of the percentage of the supply that is staked. When applying the 800 million STRK that Starknet indicated as staked in its quarterly report on November 4, 2025, it results in an annual rate of about 1.12%. This is similar to the "approximately 1%" indicated in the report. While this is a small figure, it is not zero, and there is no mechanism in the contract to stop the increase.

There is another explanation regarding the circulating supply that is omitted from the listing pages. Starknet's documentation states that tokens held by the foundation, even if unlocked by contract, are not included in the circulating supply until they are granted or distributed. Therefore, the aggregator's figures are not directly read from the chain but include judgments regarding wallet handling. This is why the circulating supply and total supply differ. Quoting one of the figures without clarifying which one is being referred to can lead to discrepancies of billions.

What Can Be Verified with Validity Proofs

It is similar to the difference between seeing the results of an exam and receiving a receipt that shows the grading was done correctly. Optimistic rollups assume that the batch is correct and provide a period for anyone to contest it. Validity rollups prove the batch first, and Ethereum verifies the proof rather than the processing itself.

Starknet's documentation clearly explains this processing flow. It generates and aggregates proofs to compress the execution of multiple blocks into a single concise piece of data. By sending that data to Ethereum for verification, Starknet's execution results can be verified without re-executing the processing. Furthermore, the compressed state differences are also published on Ethereum, allowing for independent reconstruction and verification of the complete state.

What is obtained is a confirmed process that does not require waiting for a contestation period and does not solely depend on the integrity of the overseers. On the other hand, it does not result in a network without operators. Starknet itself explains this point. The official documentation states that while responsibility is gradually transferred to validators, the chain is centralized. For more on the differences between validity proofs and optimistic rollups, you can refer to this explanation.

Assets Protecting Starknet Are Not Just STRK

Starknet's staking documentation includes parameters not mentioned on token listing pages. In the mainnet's staking power, Bitcoin has a weight of 0.25, while STRK has a weight of 0.75. Bitcoin holders lock a selected tokenized representation of Bitcoin on Starknet and receive rewards in STRK. Based on this mechanism, Starknet describes itself as the first rollup to adopt dual-token consensus. According to the mainnet rules, a minimum of 20,000 STRK is required to operate as a validator, and there is a 7-day lock period until withdrawal is completed.

Understanding the scale has its challenges beyond the design itself. Starknet reported in its quarterly report on November 4, 2025, that over 800 million STRK, more than 19% of the circulating supply, is staked. It also states that the total amount, including 659 BTC, exceeds 150 million dollars. These are figures from past points in time, and the network has continued to evolve since then. They should be treated as the most recent figures published by the issuer, not current values.

Regardless of how the current total changes, the key design points remain unchanged. A quarter of the security budget for Ethereum rollups is represented by assets that are not directly related to Ethereum. Rewards for those assets are paid in the tokens discussed in this article.

Factors That Could Affect STRK Price

Scheduled Unlocks. Starknet's documentation specifies monthly releases for investors and early contributors. Until March 15, 2027, 1.27% of the supply, approximately 12.7 million tokens, is scheduled to be released on the 15th of each month. This is not a rumor that needs to be factored into the market but a verifiable schedule.

Fee Demand. Since fees are paid only in STRK, the usage of the network could become a demand factor for STRK. It is important to verify not only the announcements but also the actual usage.

Participation in Staking. The minting curve links both. As the amount of STRK staked increases, the annual minting amount increases, and as the minting amount increases, the holdings of non-staking holders are diluted. The rate is low, so it is more of a gradual pressure than a sudden change.

Expansion of Bitcoin Staking Has Two Sides. Every time a new tokenized Bitcoin is locked for consensus, newly minted STRK is paid as a reward. Therefore, this design enhances security while also increasing the supply available for sale.

Macroeconomic Events Also Influence Price Formation. The U.S. Producer Price Index will be released on Thursday, September 10, at 12:30 UTC, and the Consumer Price Index will be released at the same time on Friday, September 11. The U.S. Federal Reserve will hold meetings on September 15 and 16 and publish economic forecasts. Small tokens may move more with these economic indicators than with their own news.

Risks

There is no supply cap by design. Regardless of what is displayed on listing pages, there is no maximum supply for STRK. A valuation model fixed at 10 billion would be based on a figure that is already off by 162.5 million tokens.

The displayed market capitalization varies depending on which supply figure is used. At the closing price of $0.030574795 on Tuesday, September 8, using the published circulating supply gives a market cap of about $21,960,000. Using the total supply from the Ethereum contract gives about $31,070,000. Even on the same price and day, the choice of denominator results in a difference of about $9,110,000.

The network is still under development. Starknet's staking documentation states that the protocol is in the second phase of a four-phase process and describes the transitioning chain as centralized. Decentralization is not a realized nature yet, and it is a goal on the roadmap.

A small group can change core contracts. Starknet describes a Security Council of 12 individuals who have the authority to manage the core contracts of Starknet and Ethereum. This council can upgrade contracts and pause or resume them. While this mechanism is in place to respond to attacks within minutes, it is important to note the concentration of authority.

There is a relatively large difference between the two price feeds. CoinGecko and CoinPaprika were 1.786% apart at the closing price on Tuesday, September 8, which was the largest difference among the assets checked during that session. For tokens traded at decimal prices, the displayed price can vary depending on which feed is referenced.

Conclusion

What is important about Starknet is not just that a data provider got one figure wrong. The same incorrect figure has been widely published, and the inaccurate state has persisted for a long time. Verification can be done with two queries to a public node that anyone can execute for free.

The supply should not be memorized as fixed information but read as the latest value to be confirmed along with the block number. The figure of 10 billion was correct as of November 30, 2022, and the block confirmed this time exceeds 10 billion. This discrepancy is the reason to verify the data yourself. This applies to all tokens for which staking rewards are set, not just STRK.