ESPIRE - Pick Execution: leaderboard
Metric: Execution acceptance rate (%) on the 1,095 pick tasks: the model's proposed gripper pose is accepted by the simulated robot within 5 attempts (a gold target replaces a failed localization); no reflection input; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 63.93 | #145 |
| 2 | Qwen 3 VL 8B Instruct | 63.2 | #401 |
| 3 | InternVL3-78B | 63.01 | #345 |
| 4 | Qwen 3 VL 30B A3B Instruct | 62.56 | #365 |
| 5 | Qwen 3 VL 235B A22B Instruct | 52.79 | #264 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=espire-pick-execution · How It Works · Data refreshed daily, snapshot 2026-10-11.