Share your thoughts, 1 month free Claude Pro on usSee more

Building Blocks on Real-world robot tasks 1.0 (test)

80Success Rate

Claude-3.7

Updated 7d ago

Evaluation Results

Method	Links
Claude-3.7 2026.05		80
Claude-4.6 2026.05		80
Qwen2.5-32B 2026.05		70
GPT-4o 2026.05		60
GPT-5 2026.05		60
Qwen3-8B 2026.05		50
Qwen3-32B 2026.05		50
Qwen3.5-35B 2026.05		50
GPT-4o 2026.05		0
GPT-5 2026.05		0
Claude-3.7 2026.05		0
Claude-4.6 2026.05		0
Qwen3-8B 2026.05		0
Qwen3-32B 2026.05		0
Qwen2.5-32B 2026.05		0
Qwen3.5-35B 2026.05		0