Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Spatial decision-making on FrozenLake size 12
Loading...
64.72
Success Rate
SpecFlow
20.676
32.1105
43.545
54.9795
Jun 1, 2026
Success Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Success Rate
SpecFlow
Setting=Flow(image) +...
2026.06
64.72
Gemini-3-Flash
Setting=Naive
2026.06
62.31
DiffThinker
Setting=Flow(image) Only
2026.06
60.41
GPT-5.1
Setting=Naive
2026.06
58.74
MVoT
Setting=Autoregressive
2026.06
58.62
Qwen3-VL-32B
Setting=SFT
2026.06
47.36
Qwen3-VL-32B
Setting=GRPO
2026.06
44.02
Qwen3-VL-8B
Setting=SFT
2026.06
37.86
Qwen3-VL-8B
Setting=GRPO
2026.06
34.56
Qwen3-VL-32B
Setting=Naive
2026.06
33.06
Qwen3-VL-8B
Setting=Naive
2026.06
22.37
Feedback
Search any
task
Search any
task