AgentRoboCritic - Failure Mode Accuracy: leaderboard

AgentRoboCritic (arXiv:2611.26613v1, Table 1): zero-shot multimodal action-level critic accuracy (%) on classifying the failure mode across 264 human-annotated failed robot-control actions (off-the-shelf models only); higher is better.

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

Top models

#ModelScoreOverall rank
1GPT-6 Astra80.68#10
2Claude Fable 5.272.73#1
3GPT-6.1 Sol62.12#15
4GPT-6.1 Terra61.36#25
5Claude Opus 5.560.98#9
6GLM-5.3 Flash42.8#133
7DeepSeek V4.1 Flash (1017)32.95#47
8Claude Sonnet 5.532.58#24
9Muse Glimmer 30B32.58#398
10Kimi K3.1 Pro32.2#36

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

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