Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Point-based Interaction on VABench-P
Loading...
73.5
Box-Hit
Embodied-R1-3B
15.0832
30.2491
45.415
60.5809
Mar 14, 2026
Box-Hit
Updated 1mo ago
Evaluation Results
Method
Method
Links
Box-Hit
Embodied-R1-3B
Model Category=Open-so...
2026.03
73.5
Gemini-3-flash
Model Category=Closed-...
2026.03
61
GPT-5.2
Model Category=Closed-...
2026.03
60.33
ST-VLA
Model Category=Ours
2026.03
59.67
Robobrain2.0-3B
Model Category=Open-so...
2026.03
26.88
Qwen3v1-4B
Model Category=Open-so...
2026.03
23.17
PEEK-3B
Model Category=Open-so...
2026.03
17.33
Feedback
Search any
task
Search any
task