BenchLeader

Robo-bench

Controlling a robot arm and a humanoid in a real-time physics simulation from motor commands and camera images. Run by Vals AI.

As of 24 Oct 2026, GPT-6.1 Sol leads Robo-bench on BenchLeader with 28.3%, ahead of GPT-6 Astra at 23.3%, across 7 model configurations with a published result.

Published by
Vals AI
Category
Agents & tools
Index weight
Reference only
Models
7
Data as of
24 Oct 2026
Newest result
13 Oct 2026

Vals AI (vals.ai).

7 of 7
#
1GPT-6.1 Solmax28.3%69.92026-10-13
2GPT-6 Astramax23.3%70.42026-10-13
3Claude Opus 5.5max14.6%72.62026-10-13
4Muse Wonder Fullmax10.4%70.72026-10-13
5Claude Fable 5.1max5.8%70.52026-10-13
6Gemini 4 Argonhigh0.8%70.82026-10-13
7Grok 4.7xhigh0.2%61.62026-10-13

More agents & tools benchmarks · All LLM benchmarks

Cite as: BenchLeader, “Robo-bench leaderboard”, data as of 24 Oct 2026, https://www.benchleader.com/benchmarks/vals_robo_bench.

Robo-bench: questions

What does Robo-bench measure?
Controlling a robot arm and a humanoid in a real-time physics simulation from motor commands and camera images. Run by Vals AI. Scores are reported in percent of tasks solved; higher is better.
Which AI model leads Robo-bench?
GPT-6.1 Sol leads Robo-bench with 28.3% as of 24 Oct 2026, ahead of GPT-6 Astra at 23.3%.
How many models have Robo-bench results?
7 model configurations have a Robo-bench result on BenchLeader, all taken from Vals AI.
Who runs Robo-bench and how often is it updated?
Robo-bench is published by Vals AI. BenchLeader re-reads the published results once a day and records the date each result was published.
Does Robo-bench count toward the BenchLeader Index?
No. Robo-bench is shown for reference but left out of the composite index.