Claude Introduces Fable 5.1 and Mythos 5.1
Anthropic positions the new models for autonomous coding and knowledge work.
Claude's official account stated it is introducing Fable 5.1 and Mythos 5.1. The post described both as the world's most advanced models for coding and knowledge work, built for complex multi-step tasks that run on their own. Users shared tables comparing Fable 5.1 benchmarks against prior versions and other models. Some highlighted gains on coding and reasoning tests. One reply praised the model while noting the Claude Code harness still has room for improvement. The announcement appeared with a video attachment.
Combined views
1.9M
29 posts, first seen 12h ago
Reactions
28 postsCH 
Chubby♨️@kimmonismusFable 5.1 benchmarls are insane. Ngl, i did not expect such jumps. Terminal-Bench 4.0, Science-Bench, HLE, crazy jumps. Now its up to OpenAI and Astra to answer that release
- likes: 2.4K
- replies: 122
- bookmarks: 187
- reposts: 126
EL elie@eliebakouchwow wait the Fable 5.1 score on terminal bench science are just totally insane??
- likes: 245
- replies: 6
- bookmarks: 25
- reposts: 11
LA Lisan al Gaib@scaling01Fable 5.1 test time compute scaling on TerminalBench and CursorBench
- likes: 49
- replies: 0
- bookmarks: 3
- reposts: 2
CL Claude@claudeaiFable 5.1 excels at complex, long-running tasks. And its research capabilities offer an early glimpse of how AI models will contribute to scientific progress.
- likes: 4.2K
- replies: 73
- bookmarks: 357
- reposts: 168
NO Noctus@noctus91Claude Fable 5.1 is looking scary. it scores: 52.6% on agentic scientific research 55.8% on Terminal-Bench 77.9% on OSWorld 73.4% on CursorBench Beating Fable 5 and GPT-5.6 Sol across multiple benchmarks.
- likes: 24
- replies: 4
- bookmarks: 0
- reposts: 3
NO Noctus@noctus91🚨 Claude Fable 5.1 is MONSTROUS. Anthropic just dropped a new frontier model that is crushing Fable 5 across coding, reasoning, computer use and agentic work. It scores: 55.8% Terminal-Bench 73.4% CursorBench 60.9% Humanity’s Last Exam 77.9% OSWorld 31.4% AutomationBench…
- likes: 14
- replies: 3
- bookmarks: 1
- reposts: 2
