Kimi K3.1 Pro scores 21.3% on humanoid BiGym 2 benchmark at 20x GPT-6 Sol's cost
teortaxesTex · x · 2026-10-26
On the BiGym 2 humanoid benchmark—where each coding agent writes a whole-body humanoid controller from one demo video, with 101,000 environment steps per session—Kimi K3.1 Pro scored 21.3%, behind GPT-6 Astra (64.5%, $11.0/task), Claude Opus 5.5 (60.6%, $15.6/task) and GPT-6.1 Sol (40.7%, just $1.3/task).
teortaxesTex notes Kimi was earlier to high-quality vision and much bigger, so it should do well on robotic/spatial/agent tasks—but it costs roughly 20x more per task ($12.1, 126k output tokens). He argues GPT-6.1 Sol must now be the north star for Chinese models.
Related event: Kimi K3.1 Pro Matches Top Tier on Robot Benchmarks at 20x Cost(3 posts)→
More from Embodied
- Open-source 3D-printed litter-picking robot MOSS tackles beach sand, gets stuck on big dunes — Scobleizer · 2026-10-26
- Visual map of 12 humanoid robot makers: valuations, flagships and production targets — CurieuxExplorer · 2026-10-26
- Matt Wolfe: robot butler demo and this week's AI news roundup — Matt Wolfe · 2026-10-26
- CoRL paper: one-step inference matches multi-step for large VLAs, 1.6-4.8x faster action generation — kaixhin · 2026-10-26
- Reverse-engineered Ray-Ban Meta glasses show chess moves in AR at 3000+ Elo — Vjeux · 2026-10-26
- HKU/ZJU/Fudan with Lightspeed launch HiSTEVE: hierarchical VLA agent playing Minecraft via native mouse and keyboard — tencent · 2026-10-26