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

Robotic Manipulation on PrimitiveSkill

100Place Success Rate

Qwen2.5-VL-72B

Updated 28d ago

Evaluation Results

Method	Links
Qwen2.5-VL-72B 2026.05		100	50	0	100	44
VAGEN-Full 2026.05		100	88	100	100	97
GLANCE-Full 2026.05		100	88	100	100	97
VAGEN-Base 2026.05		100	88	88	88	91
GLANCE-Base 2026.05		100	88	100	88	94
GLANCE w/ Turn-PPO 2026.05		100	63	0	100	66
o4-mini 2026.05		100	50	0	75	56
Gemini 2.5 Pro 2026.05		63	63	0	75	50
Claude 4.5 Sonnet 2026.05		63	50	0	100	53
Claude 3.7 Sonnet 2026.05		63	13	0	100	44
GPT-4o 2026.05		50	63	0	88	50
Qwen2.5-VL-7B 2026.05		0	0	0	75	19
Qwen2.5-VL-3B 2026.05		0	0	0	0	0
VLM-R1-3B 2026.05		0	0	0	0	0
Turn-PPO w/ Mask 2026.05		0	0	0	100	25
Vanilla-PPO 2026.05		0	0	0	0	0
GRPO w/ Mask 2026.05		0	0	0	100	25