Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-Agent Pathfinding on 5-agent evaluation set (10x10)
Loading...
8
Valid Rate
Qwen3-4B + GRPO + LLM-as-Environment-Engineer
-0.32
1.84
4
6.16
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
2026.06
8
5.33
Qwen3-4B + GRPO
training_config=random
2026.06
7.33
3.33
Grok-4.2
2026.06
6.67
5.33
GPT-5.4
2026.06
6
3.33
Kimi-K2.5
2026.06
6
2.67
Gemini-3.1-Pro
2026.06
0.67
0
Qwen3-4B
variant=base
2026.06
0
0
Feedback
Search any
task
Search any
task