InferenceBench: Claude Opus 5.5 posts 12.08x speedup, first agent to beat hyperparameter search
maksym_andr · x · 2026-09-25
The InferenceBench team reports a landmark result: Claude Opus 5.5 becomes the first AI agent in the benchmark's history to outperform hyperparameter search, with a 12.08x speedup versus 9.83x for Fable 5.1.
When InferenceBench launched, search beat every agent tested. The tables have now turned, which the team calls a milestone for agent capabilities.
Related event: Claude Opus 5.5 First AI Agent to Beat Hyperparameter Search, 12x Speedup(3 posts)→
More from Models
- Matt Shumer says Opus 5.5 spontaneously added helicopter easter eggs to his website — mattshumer_ · 2026-09-26
- One amphibian question can tell if a model treats your prompt as a capability eval — AdtRaghunathan · 2026-09-26
- Perceptron launches Mk1.5, one embodied AI model for drones, quadrupeds and smart glasses — code_star · 2026-09-26
- Tester: Astra is "autistic" at parsing human sentiment; Fable and Opus run circles around it — teortaxesTex · 2026-09-26
- User generates an anime-style fight scene entirely with code using Claude Opus 5.5 — EricBuess · 2026-09-26
- Dev verdict on Opus 5.5: the first good Opus since 4.8 — holdenmatt · 2026-09-26