Steve Yegge: Two Weeks With Opus 5.5 — Precision Rivals Fable, Recall Trails on Open-Ended Tasks
Steve_Yegge · x · 2026-10-03
Veteran developer Steve Yegge shares his experience after two weeks with Opus 5.5, calling it an awesome model that he strongly prefers over Fable for explanations and teaching.
His custom evals show:
- Precision: Opus 5.5 rivals Fable 5.1 across all tasks in his system
- Recall: not as good as Fable
- Specific tasks: the two are roughly equal
- Open-ended tasks: Fable more often spots and acts on important issues, showing better perspective
He still considers Opus 5.5 an incredible model and now uses it for the majority of his work.
More from Models
- User Claims 'GPT-6 Astra Dots' Built a Full 3D Palace in Blender Autonomously — 141_1337 · 2026-10-03
- Why SFT generalizes worse than RL: off-policy data, not the objective — a_karvonen · 2026-10-03
- Opus too pricey at high effort, not meaningfully better than GPT-6 astra — haider1 · 2026-10-03
- User asks Grok about a noise overhead — the agent opens a browser and tracks the helicopter — mertdumenci · 2026-10-03
- Critic questions Tavus's AI human claims: fails Turing test, unavailable to test — churchkey · 2026-10-03
- Xiaomi's MIT-licensed MiMo-V2.6-Pro-RL tops open-weights intelligence index, discloses ~$2.6M RL training cost — lmoroney · 2026-10-03