Multi-Agent Pathfinding on 5-agent 5x5 (evaluation set)
28Valid RateQwen3-4B + GRPO + LLM-as-Environment-Engineer
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Qwen3-4B + GRPO + LLM-as-Environment-Engineer2026.06 | 28 | 18 | |
| Grok-4.22026.06 | 26.67 | 16 | |
| Qwen3-4B + GRPOtraining_config=random2026.06 | 24 | 16.67 | |
| Kimi-K2.52026.06 | 23.33 | 17.33 | |
| GPT-5.42026.06 | 17.33 | 14 | |
| Gemini-3.1-Pro2026.06 | 9.33 | 8 | |
| Qwen3-4Bvariant=base2026.06 | 2.67 | 2 |