Benchmarks · agentic
AA-Robotics
AA-Robotics: a model in its lab's coding-agent harness directly controls a 6-axis robot arm (5 joints and a gripper) through everyday tabletop tasks such as pressing a button, pouring a drink and writing "hello world". Mean score per task (0 to 100, partial credit for progress).
5 models · 5 results · data as of 2026-10-29 · artificialanalysis.ai ↗
Leaderboard
GPT-6 Astra high
More from this benchmark
6 published measuresScore vs any published measure6 measures: sub-scores, splits, counts
Breakdown
darker is better within each column · click a column to sort| Model | ||||||
|---|---|---|---|---|---|---|
| GPT-6 Astra high | 100.0 | 40.0 | 44.0 | 100.0 | 71.0 | 0 |
| GPT-6.1 Sol high | 0 | 100.0 | 100.0 | 100.0 | 0 | 17.0 |
| Gemini 4 Argon high | 100.0 | 40.0 | 0 | 45.0 | 57.0 | 0 |
| Opus 5.5 high, fallback | 25.0 | 40.0 | 38.0 | 45.0 | 25.0 | 17.0 |
| Fable 5.1 high, fallback | 30.0 | 40.0 | 100.0 | 0 | 0 | 0 |
All results
5 rows| # | Model | Effort | AA-Robotics | Harness |
|---|---|---|---|---|
| 1 | GPT-6 Astra | high | 69.4 | Codex |
| 2 | GPT-6.1 Sol | high | 64.6 | Codex |
| 3 | Gemini 4 Argon | high | 55.3 | Antigravity |
| 4 | Opus 5.5 | high | 48.8 | Claude Code |
| 5 | Fable 5.1 | high | 46.3 | Claude Code |