Long-context Memory Retrieval and Reasoning on PersonaMem 32K
26.45F1 ScoreMemSifter
Evaluation Results
| Method | Links | |
|---|---|---|
| MemSifterBackbone LLM=Qwen3-30B-A3B-Ins2026.03 | 26.45 | |
| BGE-M3Backbone LLM=Qwen3-30B-A3B-Ins2026.03 | 24.98 | |
| Mem0Backbone LLM=Qwen3-30B-A3B-Ins2026.03 | 24.87 | |
| A-MEMBackbone LLM=Qwen3-30B-A3B-Ins2026.03 | 24.57 | |
| NemoriBackbone LLM=Qwen3-30B-A3B-Ins2026.03 | 24.35 | |
| Qwen3-30BBackbone LLM=Qwen3-30B-A3B-Ins2026.03 | 24.25 | |
| ReasonRankBackbone LLM=Qwen3-30B-A3B-Ins2026.03 | 23.78 | |
| MemSifterBackbone LLM=DeepSeek V3.22026.03 | 23.7 | |
| HippoRAGBackbone LLM=Qwen3-30B-A3B-Ins2026.03 | 23.35 | |
| GemmaEmbBackbone LLM=Qwen3-30B-A3B-Ins2026.03 | 23.3 | |
| HippoRAGBackbone LLM=DeepSeek V3.22026.03 | 22.65 | |
| RearankBackbone LLM=Qwen3-30B-A3B-Ins2026.03 | 22.37 | |
| Mem0Backbone LLM=DeepSeek V3.22026.03 | 22.33 | |
| RearankBackbone LLM=DeepSeek V3.22026.03 | 22.23 | |
| DS-V3.2Backbone LLM=DeepSeek V3.22026.03 | 21.69 | |
| ReasonRankBackbone LLM=DeepSeek V3.22026.03 | 21.6 | |
| A-MEMBackbone LLM=DeepSeek V3.22026.03 | 21.44 | |
| NemoriBackbone LLM=DeepSeek V3.22026.03 | 21.25 | |
| GemmaEmbBackbone LLM=DeepSeek V3.22026.03 | 20.96 | |
| BGE-M3Backbone LLM=DeepSeek V3.22026.03 | 20.76 |