Multi-Agent Pathfinding on 5-agent Sum (test)
18.67Valid RateQwen3-4B + GRPO + LLM-as-Environment-Engineer
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Qwen3-4B + GRPO + LLM-as-Environment-Engineer2026.06 | 18.67 | 11 | |
| Qwen3-4B + GRPOtraining_config=random2026.06 | 15.11 | 9.11 | |
| Kimi-K2.52026.06 | 13.47 | 8.78 | |
| Grok-4.22026.06 | 12.89 | 8.78 | |
| GPT-5.42026.06 | 9.11 | 6 | |
| Gemini-3.1-Pro2026.06 | 4.78 | 3.56 | |
| Qwen3-4Bvariant=base2026.06 | 1.44 | 1.22 |