GPT-6.1 Sol tops robotics evals at 19.1%, beating Anthropic's Opus 5.5
OpenAIDevs · x · 2026-10-20
Eval lab XDOF published overall task success rates for four frontier models across 16 shared robotics tasks (1,868 scored runs, 95% bootstrap CIs):
- GPT-6.1 Sol: 19.1% (473 runs, CI 16.8–21.4%)
- GPT-6 Astra: 17.2% (464 runs, CI 14.3–20.2%)
- Opus 5.5: 11.6% (458 runs, CI 9.4–13.8%)
- Fable 5.1: 9.1% (473 runs, CI 7.5–11.0%)
Both OpenAI models rank on top, with Sol outperforming Astra and both clearly ahead of their Anthropic counterparts.
More from Models
- OpenAI's Decisions API hits GA: near real-time routing up to 10x faster than GPT-6 Luna — OpenAIDevs · 2026-10-23
- DeepLearning.AI weekly: GPT-6.1 hits 97.9% byte parity on COBOL migration — DeepLearningAI · 2026-10-23
- Marker's lite mode scores 75.1 vs 42.3 for Azure Read on OmniParseBench — VikParuchuri · 2026-10-23
- Datalab launches lite OCR at $0.50 per 1,000 pages, 20-30x cheaper than Azure, Google and AWS layout APIs — VikParuchuri · 2026-10-23
- LiveNerf's 29-day data finds no nerfing of Claude Opus 5.5 — TheOnlyVibemaster · 2026-10-23
- PHP Foundation reveals Anthropic's Claude cyber model scanned PHP, validated with OpenAI's Daybreak Red — saibharadwaj · 2026-10-23