Rumors Say OpenAI's Internal Model Cracks Hodge and BSD; Bel Benchmark Near 50%
adonis_singh · x · 2026-09-11
Analyzing rumors that OpenAI's internal model solved the Hodge and Birch–Swinnerton-Dyer conjectures, @adonissingh notes the model "Bel" scores 50% on OpenAI's curated set of open math problems despite starting post-training around Aug 28, while the strongest public math model Astra gets <20% on hard problems. He cautions the rumors could be baloney but argues such results are coming soon regardless.
More from Models
- MissionReport benchmark: only 3 of 540 AI-generated government reports pass all criteria — Exp_Mark · 2026-10-27
- Local beats cloud: Qwen 3.8 FN handles private student data as well as frontier models — dr_lm · 2026-10-27
- Aidan Gomez: model's CoT said no tool calls right before making them — aidangomez · 2026-10-27
- Nat Lambert: Cohere's model is narrow RL training on internal infra ill-suited for open weights — natolambert · 2026-10-27
- OpenAI teases DOTS is coming to Europe 'soon' — mark_k · 2026-10-27
- Clifford Pickover challenges AI to decipher a message he encoded in Norse runes — pickover · 2026-10-27