Fable 5.2 hits 174 on Epoch Capabilities Index, beating GPT-6 Astra's 169
Miles_Brundage · x · 2026-11-10
Epoch AI reports that Claude Fable 5.2 set a new record of 174 on its Capabilities Index (ECI), surpassing GPT-6 Astra's previous high of 169 and beating the pre-Astra trendline for reasoning models. Epoch ties this to recent discussion of AI contributions to AI R&D and plans to release more work soon. Scores are based on pre-launch testing with Epoch's own benchmarks and may change as the ECI updates.
Related event: Claude Fable 5.2 Sets ECI Record of 174, Topping GPT-6 Astra(4 posts)→
More from Fun
- AI ethics spat: Lorenz lecturing journalists mocked over BuzzFeed plagiarism firing — ShakeelHashim · 2026-11-10
- Titanic passengers with glowing AI laptops: the meme that hits too close to home — generativist · 2026-11-10
- GPT 6.1 agent designs a cascadable broadband half-adder in a 23-minute run — CatAstro_Piyush · 2026-11-10
- Imagine a world where Claude was called Norbert — names shape model personas — matt_slotnick · 2026-11-10
- "What the hell is a 'slice' of an app?" — dev mocks AI coding jargon — vboykis · 2026-11-10
- fal team shows off AI-generated custom brick totem made via 2ndbrickstudio — noahsolomon · 2026-11-10