Claude Fable 5.1 Doubles Scores on Science Benchmarks
testingcatalog · x · 2026-09-02
Anthropic announced Claude Fable 5.1 and Claude Mythos 5.1. The model scores 52.6% on Terminal-Bench-Science 0.1, more than doubling Fable 5's performance. On Terminal-Bench 4.0, it achieves 55.8% compared to Fable 5's 42.0%. Features include plain language writing, spreadsheet creation with number verification, and clear source citation.
Related event: Anthropic Releases Fable 5.1 and Mythos 5.1 with 75% Cheaper Caching(174 posts)→
More from Models
- Mocked benchmark shows DeepSeek's document chart grounding badly misaligned — burkov · 2026-10-18
- Dev complains Claude's habit of saying "not done" is his biggest pain point — jasonkneen · 2026-10-18
- Open-weight TYGER speech decision model makes 80ms turn-taking calls for full-duplex voice agents — solyarisoftware · 2026-10-18
- Paying Google AI subscriber blocked: custom-domain Workspace accounts can't access AI Studio — ChrisRackauckas · 2026-10-18
- Running Ling 3.1 Flash (560B MoE, 25B Active) Across 4 DGX Sparks via llama.cpp RPC — BerryBlush_Rogers691 · 2026-10-18
- Users press OpenAI on long-delayed Adult Mode a year after it was promised — ___Patrice___ · 2026-10-18