Robo-bench (Vals AI) leaderboard
- GPT-6.1 Sol leads
- 28.3%
- models measured
- 6
- lead over #2
- +5.0 pts
- median
- 10.2%
Leader on Robo-bench (Vals AI) · read October 22
On October 22, 2026, 6 of our catalog's models had a Robo-bench (Vals AI) result, and OpenAI GPT-6.1 Sol led with 28.3%, ahead of OpenAI GPT-6 Astra at 23.3%.
Control of a robot arm and a humanoid in a real-time physics simulation, from motor commands and camera images alone. The score as Vals AI publishes it.
Role in the index
Robo-bench (Vals AI) is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | GPT-6.1 SolOpenAI | 28.3% | 82.4 | $10.00 | |
| 2 | GPT-6 AstraOpenAI | 23.3% | 85.1 | $50.00 | |
| 3 | Claude Opus 5.5Anthropic | 14.6% | 88.8 | $20.00 | |
| 4 | Claude Fable 5.1Anthropic | 5.8% | 84.2 | $50.00 | |
| 5 | Gemini 4 ArgonGoogle | 0.8% | 81.0 | $10.00 | |
| 6 | Grok 4.7xAI | 0.2% | 66.7 | $6.00 |
Only models in our catalog are listed, and each appears once with its best published run. 137 rows were read from the source on October 22, 2026, and rows that match no catalog model are left out.