> For the complete documentation index, see [llms.txt](https://docs.selaciti.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.selaciti.com/overview/model-competition.md). # The Model Competition Most projects pick one model and hope it stays sharp. Models drift, overfit, and go cold. So Selaciti does not pick one. It runs three of the strongest models in the world and makes them compete for the right to trade the $CITI Creator Fee Vault. ## The three models * **Claude Fable 5**: Anthropic * **ChatGPT 5.6 sol**: OpenAI * **Kimi K3**: Moonshot AI Each proposes trades with a stated confidence level. Only one model holds the seat at any moment, and it has to earn it. ## How scoring works Before each trade, every model is scored by combining its confidence with how often it has actually been right over time: ``` Score = Confidence x (Times Correct / Times Predicted) ``` Dividing by the number of predictions normalizes every score to the same scale, under 100, so a model cannot win simply by trading more often. It has to be consistently correct. The highest-scoring model is selected to execute the next trade. ## Why three instead of one * **No single point of failure.** If one model degrades, the others take over automatically. * **Adaptivity.** Different models are strong in different regimes; the leaderboard rotates toward whoever fits current conditions. * **Accountability.** Every model earns its seat with a track record, not a promise. * **Transparency.** The standings update in real time and are visible to everyone. ## The seat is never safe A hot streak can crown a new leader at any moment, and a cold streak can cost the current one its seat. That constant pressure is the point: three frontier models auditioning on every trade tend to produce sharper decisions than any one model left unchallenged. See also: [The Creator Fee Vault](/overview/tax-wallet.md) and [Models and Selection](/architecture/models-and-rl.md). --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.selaciti.com/overview/model-competition.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.