OpenAI May Already Be Pretraining GPT-6.5 That Beats Astra on Evals
willdepue · x · 2026-09-09
willdepue parses OpenAI's wording that its Navier-Stokes agents used a model "significantly more capable than GPT-6 Astra" and concludes:
- OpenAI has likely already started astra-next / GPT-6.5 pretraining
- It is already beating Astra on evals
- Suggests OpenAI's model cadence is far ahead of public expectations.
Related event: Engineer Speculates OpenAI Has Started GPT-6.5 Pretraining, Beating Astra(2 posts)→
More from Models
- Researcher says he predicted Navier-Stokes would fall first, but not amid bitter controversy — geoffreyirving · 2026-09-09
- 'Context-window AGI': inside a single window, models beat any individual — sir_codes_alot · 2026-09-09
- littmath explains why models cracked Navier-Stokes but Riemann is harder — littmath · 2026-09-09
- OpenAI's Internal Model Cracks Open Math Problems Where Humanity Scored Effectively Zero — stevenheidel · 2026-09-09
- GPT-6 Astra Tops Design Arena's 3D Leaderboard With Elo 1495, 65 Pts Ahead of Kimi K3 — airesearch12 · 2026-09-09
- If the NS proof took humans years, internal models may reach decades-long METR horizons — Realistic_Stomach848 · 2026-09-09