Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Failure Detection on RoboFail
Loading...
67
Accuracy
Gemini 2.0 Flash
50.36
54.68
59
63.32
Mar 16, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Gemini 2.0 Flash
Model Source=Closed-So...
2026.03
67
PRIMO R1
Model Source=Ours
2026.03
67
Cosmos-Reason1-56B
Model Source=Open-Source
2026.03
66.2
Nemotron-H-56B
Model Source=Open-Source
2026.03
64
GPT-4o
Model Source=Closed-So...
2026.03
63
PRIMO (RL)
Model Source=Ours, Tra...
2026.03
63
OpenAI o1
Model Source=Closed-So...
2026.03
61
Cosmos-Reason1-7B
Model Source=Open-Source
2026.03
60
Claude-haiku-4.5
Model Source=Closed-So...
2026.03
59
Qwen2.5-VL-7B
Model Source=Open-Source
2026.03
57.6
PRIMO (SFT)
Model Source=Ours, Tra...
2026.03
51
Feedback
Search any
task
Search any
task