AgentRoboCritic - Overall Accuracy: leaderboard

AgentRoboCritic (arXiv:2611.26613v1, Table 1): zero-shot multimodal action-level critic evaluation on 400 human-annotated direct robot-control actions from 12 manipulation benchmarks. Overall accuracy (%) requiring both action outcome and failure mode correct (off-the-shelf models only); higher is better.

Metric: Overall Accuracy (%). Source: arxiv.org. Saturation forecast: Around May 2027. 10 models tracked. Updated 21 Nov 2026.

Top models

#ModelScoreOverall rank
1GPT-6 Astra71.25#10
2Claude Fable 5.271#1
3GPT-6.1 Sol69.75#15
4GPT-6.1 Terra66.75#25
5Claude Opus 5.566#9
6GLM-5.3 Flash58.25#133
7Kimi K3.1 Pro52.5#36
8DeepSeek V4.1 Flash (1017)52#47
9Claude Sonnet 5.550.75#24
10Muse Glimmer 30B34.75#398

No result here: #3 Gemini 4 Neon, #4 GPT-6.2 Astra, #7 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=agentrobocritic-overall-accuracy · How It Works · Data refreshed daily, snapshot 2026-11-21.