Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-Agent Path Finding on MAPF-FrozenLake 4-agent 10x10
Loading...
18.67
Valid Rate
Qwen3-4B + GRPO + LLM-as-Environment-Engineer
-0.7468
4.2941
9.335
14.3759
Jun 16, 2026
Valid Rate
Optimal Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Valid Rate
Optimal Rate
Qwen3-4B + GRPO + LLM-as-Environment-Engineer
Backbone=Qwen3-4B, Tra...
2026.06
18.67
8
Grok-4.2
2026.06
14.67
7.33
Qwen3-4B + GRPO (random)
Backbone=Qwen3-4B, Tra...
2026.06
14
5.33
Kimi-K2.5
2026.06
9.33
6
GPT-5.4
2026.06
4
2.67
Gemini-3.1-Pro
2026.06
3.33
2.67
Qwen3-4B (base)
Backbone=Qwen3-4B, Sta...
2026.06
0
0
Feedback
Search any
task
Search any
task