GPT-6 Astra tops Epoch Capabilities Index but Claude Fable 5.1 still leads software engineering
Jsevillamol · x · 2026-09-17
Epoch AI's latest Capabilities Index (ECI) shows GPT-6 Astra taking the overall lead ahead of rivals like Claude Fable 5.1, with its Math-ECI score setting a new record.
- On software engineering benchmarks, however, Claude Fable 5.1 remains state-of-the-art; Astra is not clearly ahead there and may even trail.
- Reposter Alex Barry agrees with that profile: clearly ahead in math/reasoning/game playing, but not clearly ahead in software engineering.
Related event: GPT-6 Astra Tops Epoch Capabilities Index Despite Score Revision(2 posts)→
More from Models
- Jason Wei's Stanford talk: intelligence is becoming a commodity as adaptive compute takes off — dotey · 2026-09-18
- GPT-6-Astra beats Fable-5.1 at RollerCoaster Tycoon 2 in 3 hours, using 5x fewer tokens — scaling01 · 2026-09-18
- RL agents invent their own diagnostic renderings to ground code understanding, sparking RL scaling optimism — teortaxesTex · 2026-09-18
- Dev discovers Codex security hardening switched persistent agent sessions to per-message instances — RileyRalmuto · 2026-09-18
- Astra for Law posts big legal benchmark gains as Mollick asks if labs will eat every AI vertical — emollick · 2026-09-18
- Anthropic's stealth model accused of hardcoded routing to Opus 5 — teortaxesTex · 2026-09-18