List your company

Embodied Agent Task Completion on MCU task suite

100Wooden Success Rate (SR)Human-level
22.634442.719762.80582.8903Mar 13, 2026

Evaluation results

MethodLinks
Human-level2026.03100100100100100100100100
Steve-EvolvingLLM Backbone=Gemini-3-pro2026.0398.9794.1957.0714.6435.1215.7728.3653.37
Steve-EvolvingLLM Backbone=Qwen3.5-plus2026.0398.7193.4855.8313.7931.2417.0627.6352.52
Optimus-1LLM Backbone=Gemini-3-flash2026.0398.791.7645.926.8524.4813.0618.9846.73
Steve-EvolvingLLM Backbone=Gemini-3-flash2026.0398.6393.3755.2813.5233.7312.7827.1852.04
Optimus-1LLM Backbone=Qwen3.5-flash2026.0398.392.7545.248.0728.178.2114.845.83
Optimus-1LLM Backbone=Gemini-3-pro2026.0398.2894.346.418.5825.0914.9319.4347.63
Steve-EvolvingLLM Backbone=Qwen3.5-flash2026.0398.0992.6152.4310.1130.7714.5823.4750.09
Optimus-1LLM Backbone=Qwen3.5-plus2026.0397.9193.646.067.0532.0311.4219.0947.42
Steve-EvolvingLLM Backbone=GLM-4.72026.0397.3691.1850.558.6329.0810.9222.0748.43
Optimus-1LLM Backbone=GLM-4.72026.0396.6391.444.137.7923.067.9518.0445.23
Jarvis-1LLM Backbone=Gemini-3-pro2026.0393.5789.0935.9311.2723.499.2115.9442.67
Jarvis-1LLM Backbone=Qwen3.5-plus2026.0393.0888.7335.5711.1226.189.0315.5842.59
Jarvis-1LLM Backbone=Gemini-3-flash2026.0392.9988.5835.219.7922.968.9215.5242.04
Jarvis-1LLM Backbone=Qwen3.5-flash2026.0392.3788.1434.7610.8323.418.3914.9241.75
Jarvis-1LLM Backbone=GLM-4.72026.0391.7986.8733.948.8721.828.0814.1940.79
Steve-12026.0325.6118.722.4700007.26