Which company has the best AI model on LiveBench (Mathematics) end of September?: how this market works

What you need to know

This market is asking a simple question: whose AI model scores highest on a math benchmark called LiveBench by the end of September 2026? LiveBench is a public leaderboard that tests AI models on real math problems and scores them objectively. A win for OpenAI means one of their models sits at the top of that math column. A win for Anthropic or ByteDance means one of their models beat everyone else on that specific score. On September 30, 2026, at 12:00 PM Eastern Time, someone will check the Mathematics column on livebench.ai and read off the top score. Whichever company owns that top-scoring model wins. If two models tie on score, the cheaper one wins. If they still tie, company names are sorted alphabetically and the first one wins. The leaderboard itself is the only source that matters, so the result depends entirely on what that page shows at that exact moment. None of the recent headlines provided are relevant to this market. They cover stock markets, Amazon earnings, and unrelated topics. The kinds of updates that would actually matter here are new AI model releases from OpenAI, Anthropic, or ByteDance, or changes to the LiveBench leaderboard rankings in the mathematics category. This market is genuinely hard to predict because the AI field moves fast and a single new model release can completely reshuffle a leaderboard overnight. The market currently prices OpenAI around 49% and Anthropic around 33%, so neither is dominant. Over a year remains until the deadline, which is a very long time in AI development. Any of the companies listed, or even a surprise entrant classified as 'Other', could release a stronger math model between now and September 2026.

The odds right now

• OpenAI49%
• Anthropic32%
• Amazon7%
• Z.ai6%
• Google5%
• ByteDance5%
• Meituan5%
• Meta5%
• SpaceXAI5%
• MiniMax5%
• Moonshot4%
• DeepSeek4%

Price history

How this resolves

Resolves September 30, 2026

This market will resolve according to the company which owns the model with the highest Mathematics score on LiveBench.ai when the leaderboard is checked on September 30, 2026, at 12:00 PM ET. Results from the “Mathematics” column of the leaderboard at https://livebench.ai/#/?cats=Mathematics, with the latest available LiveBench release selected and the category set to “Mathematics,” will be used to resolve this market. Read the full resolution rules on the live market page.

Related

Same markets. A fraction of the fee.

These apps all route to the same exchange order book. The difference is what each one adds on top of the exchange's own fee.

Published rates, checked July 2026. MetaMask charges a flat 4 percent per prediction trade. Jupiter adds a fee equal to the exchange's own taker fee at fill time, roughly 2 to 4 percent at typical odds. Where a market carries an exchange settlement fee, it applies everywhere, whichever app you use. Paridesk adds nothing on maker orders.

Trade this market on Paridesk: non-custodial, 0.5% fee.