OpenAI publishes 722 math manuscripts from internal model, with proofs and reasoning traces
danielrock · x · 2026-10-07
OpenAI has released the openai/math GitHub repo containing 722 mathematical manuscripts and proof artifacts produced by an internal model. Key points:
- After existing math evals saturated, OpenAI expanded model evaluations to open research problems.
- The repo includes Lean formalizations, preprints, and reasoning traces, plus an overview PDF.
- Some unformalized results may contain issues; OpenAI says it will fix them quickly and keep adding Lean proofs, and is exploring community-hosted mirrors.
- Apache-2.0 licensed; the repo hit 1.3k stars within a day.
It's the first systematic public release of frontier-model original math research, notable for both mathematicians and capability evaluation.
More from Models
- Is Bel's math edge scale or synthetic data? TeortaxesTex bets on data — teortaxesTex · 2026-10-07
- Fable claims its new release equals roughly five OpenAI Navier-Stokes-level results — willdepue · 2026-10-07
- Reddit user ships 'surgical abliterated' 27B red-team model with zero refusals — Least_Dog_8556 · 2026-10-07
- OpenAI Researcher Surprised AI Lab Math Results So Far All Hold Up — willdepue · 2026-10-07
- OpenAI dots losing to Meta's Muse surprises AI community — BLUECOW009 · 2026-10-07
- Inception launches Mercury Decide on OpenRouter: free structured-decision model doing 14 decisions/sec — StefanoErmon · 2026-10-07