Autonomys is crafting a token issuance model that will allow our network to thrive for decades to come. We’ve invested a great deal of resources into diligently conducting a process of mainnet modeling and issuance parameter selection to this end — in collaboration with leading complex systems engineering firm BlockScience. The original analysis provided by the BlockScience team is available here. We are sharing some of our research and development findings to offer greater clarity about the conclusions we reached.
We designed the economic parameters of our protocol issuance to align with the broader objectives of the network:
The total ATC supply cap will be 1 billion, as per our simulation. The exact allocations to the team, investors and others will be announced when we have the final numbers following the conclusion of the incentivized testnet and Stake Wars 2. Our implementation keeps track of the total issued tokens to date and ensures the total supply cap is never exceeded.
The algorithm for the Token Generation Event (TGE) is largely the same as that for the Gemini-3h testnet described here (TL;DR: we will hold a Space Race that starts automatically at genesis block until a certain amount of pledged space is reached, at which point protocol rewards begin).
The results of the simulation suggest rewards are likely to start within 7 days of our mainnet launch. It did not estimate the amount of pledged space required. We will compute this in the final days of testnet based on the total space pledged, as well as the speed of growth (~1.3 PiB/day so far). This will all inform how we choose our Space Race target. The longer it runs, the longer it would take to achieve optimal community-ownership. However, a longer Space Race would slightly increase the lifetime of the protocol issuance.
NOTE: The release of our official GPU plotter (if ready before mainnet launch) will also likely impact our Space Race expectations.
The first block after the Space Race target is reached will immediately start minting tokens to reward the proposer and voters. The value of the reward issued is indicative: if we set it to $x$ ATC, the real issued value may be higher or lower depending on utilization and the number of votes, but on average, it follows the curve below. This value, which we call reference subsidy, decreases from the next block following an exponential curve (explained further below).
This issuance curve uses example values consistent with overall recommendations, and as such, these may not be final. The illustrated value per block is the total rewards distributed among all recipients in that block (as described below). Greater rewards significantly aid the growth of the community-owned supply, but at the expense of runway and distributional equilibrium (as a smaller number of early farmers receive a lot more than later ones).
Each block, on average, includes 9 votes, so we anticipated 10 recipients per block (this number may change once our sharding upgrade is implemented). Currently, the protocol issues 10% of the initial reward to the proposer, and 10% to each voter. In an empty block with exactly 9 votes, we would issue exactly 5 ATC; 5.5 ATC if there were 10 votes; 4.5 ATC with 8 votes, and so on. Assuming an initial reference subsidy of 5 ATC, the rewards for the first block would be issued as follows:
The proposer also takes a 10% tax from each voter to disincentivize them from censoring votes. The final reward per block credited to the wallets would therefore be:
The proposer share % has not shown conclusive correlation with any of the simulation goals.
To incentivize early adoption and community-owned supply growth, we have developed an issuance curve that begins with a period of slower decline, before accelerating at a certain inflection point. This is achieved by taking the sum of two exponential components: one where the decline starts immediately, and another where the decline starts after a while:
The plot shows the curve for ~15 years (in blocks), with an inflection point at 1.5 years. Although the simulation suggested the duration of this period of slower decline should be 1.5 years, it showed no strong correlation with our goals. A later inflection point would moderately benefit community-owned supply growth, but negatively impact the runway of token issuance.
For each curve, we controlled the initial reference subsidy (the starting point on the y-axis). The rate of decline is effectively controlled by the amount of tokens we ‘allocate’ to each curve component. Taking our estimated supply on TGE and allocating 50% to each component, we can expect a 40+ year runway.
The model simulated the first 3 years after TGE, and thus this runway calculation is based on the functional form of the issuance curve and the initial reference subsidy. The component allocation value is the strongest estimate for runway: lower values make the decline faster initially, but significantly prolong the tail after the four-decade mark.
The component allocation cap is not enforced in our implementation. The only enforced cap is the total supply cap. The protocol will continue issuing small amounts (i.e. ~0.1 ATC/block, as the exponential function never reaches 0) until the total supply cap is reached with no outside intervention. This allows us to gracefully handle uncertainty over the duration of the incentivized testnet as no extra tokens will be left unissued, nor will the protocol exceed the 1B cap.
Want to farm or build on the Autonomys Network?
Check out the Autonomys Academy, join the Discord & come say HAI!
Thanks for joining us on this brief tour of Autonomys Network’s mainnet protocol issuance parameters. We hope that you enjoyed this breakdown of our process, and that it helps the community better understand how we arrived at these values. Keep a lookout for our forthcoming follow-up post on protocol economics.
NOTE: The Subspace Foundation reserves the right to adjust the Autonomys Network’s token issuance parameters based on further research and testing up until mainnet launch. The Foundation does not intend for ATC tokens to be available for sale or resale within the United States or to U.S. persons.
