Claude Fable 5 Tackles Ramanujan Math Contest, Solves 3 Out of 4 on First Try
kylekabasares · x · 2026-07-04
A user tested Claude Fable 5 on the Ramanujan AI Math Challenge, solving 3 out of 4 problems on the first attempt (a 75% hit rate). The 1 unsolved problem happened to be one of the two public unsolved puzzles (a 50% hit rate on public unsolved problems). The results are still pending official verification. The challenge targets the unsolved mathematical puzzles left by Ramanujan, and Fable 5's performance has attracted significant attention.
More from Models
- ICML 2026 position paper: LLM 'thinking traces' are a harmful anthropomorphism — rao2z · 2026-10-27
- One stray 'u' and Claude starts spelling like a Brit — cantrell · 2026-10-27
- Cohere's Invent-A-Dataset paper: zero-seed synthetic data beats frontier APIs on quality and diversity — idanbeck · 2026-10-27
- ARC Prize redesigns leaderboard: highest reasoning effort by default, CSV export — mhmazur · 2026-10-27
- Cohere: open-weight Qwen3.8-Flash-Next tried uploading probe file to Beijing cloud bucket in sandbox — wstone_bd · 2026-10-27
- Model refuses wrong psychiatric advice in deliberately misleading test — FellMentKE · 2026-10-27