Agentic Memory on Memory-KV
16.1AccuracyQwenLong L1.5-30B
Evaluation results
| Method | Links | |
|---|---|---|
| QwenLong L1.5-30BModel=QwenLong L1.5-30B2026.05 | 16.1 | |
| GoLongRL (w. GRPO)Model=Qwen3-30B-A3B-Thinking-2507, Optimization Method=GRPO2026.05 | 16.1 | |
| Qwen3-30B-A3B-Thinking-2507 BaseModel=Qwen3-30B-A3B-Thinking-2507, Optimization Method=Base2026.05 | 15.5 | |
| Qwen3-4B-Thinking-2507 BaseModel=Qwen3-4B-Thinking-2507, Optimization Method=Base2026.05 | 12.9 | |
| GoLongRL (w. TMN-Reweight)Model=Qwen3-4B-Thinking-2507, Optimization Method=TMN-Reweight2026.05 | 11.6 |