Dialogue-Style Memory Reasoning on PersonaMem
37.19Exact Match (EM)REAL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| REALModel=DeepSeek-V3, Memory Category=Graph-Based Memory Baselines2026.06 | 37.19 | 49.21 | |
| REALModel=LLaMA-3.3-70B-Instruct, Memory Category=Graph-Based Memory Baselines2026.06 | 35.98 | 48.28 | |
| REALModel=Qwen3-32B, Memory Category=Graph-Based Memory Baselines2026.06 | 34.32 | 45.05 | |
| A-MEMModel=DeepSeek-V3, Memory Category=Graph-Based Memory Baselines2026.06 | 32.58 | 45.45 | |
| A-MEMModel=Qwen3-32B, Memory Category=Graph-Based Memory Baselines2026.06 | 31.34 | 42.86 | |
| Mem0gModel=LLaMA-3.3-70B-Instruct, Memory Category=Graph-Based Memory Baselines2026.06 | 31.19 | 43.25 | |
| Mem0gModel=DeepSeek-V3, Memory Category=Graph-Based Memory Baselines2026.06 | 31.04 | 43.25 | |
| Mem0Model=DeepSeek-V3, Memory Category=Flat-Text-Based Memory Baselines2026.06 | 30.14 | 42.46 | |
| A-MEMModel=LLaMA-3.3-70B-Instruct, Memory Category=Graph-Based Memory Baselines2026.06 | 30.06 | 42.8 | |
| Mem0gModel=Qwen3-32B, Memory Category=Graph-Based Memory Baselines2026.06 | 28.54 | 40.42 | |
| Mem0Model=LLaMA-3.3-70B-Instruct, Memory Category=Flat-Text-Based Memory Baselines2026.06 | 28.42 | 41.06 | |
| Vanilla Memory GraphModel=DeepSeek-V3, Memory Category=Graph-Based Memory Baselines2026.06 | 27.36 | 38.98 | |
| MemGPTModel=DeepSeek-V3, Memory Category=Flat-Text-Based Memory Baselines2026.06 | 25.87 | 37.56 | |
| Vanilla Memory GraphModel=LLaMA-3.3-70B-Instruct, Memory Category=Graph-Based Memory Baselines2026.06 | 25.74 | 37.52 | |
| Mem0Model=Qwen3-32B, Memory Category=Flat-Text-Based Memory Baselines2026.06 | 25.02 | 38.5 | |
| Vanilla Memory GraphModel=Qwen3-32B, Memory Category=Graph-Based Memory Baselines2026.06 | 24.22 | 36.4 | |
| MemGPTModel=LLaMA-3.3-70B-Instruct, Memory Category=Flat-Text-Based Memory Baselines2026.06 | 24.2 | 35.73 | |
| Flat Memory ChunkModel=DeepSeek-V3, Memory Category=Flat-Text-Based Memory Baselines2026.06 | 24 | 35.05 | |
| MemGPTModel=Qwen3-32B, Memory Category=Flat-Text-Based Memory Baselines2026.06 | 22.95 | 34.25 | |
| Full ContextModel=DeepSeek-V3, Memory Category=Full Context2026.06 | 22.52 | 33.1 | |
| Flat Memory ChunkModel=LLaMA-3.3-70B-Instruct, Memory Category=Flat-Text-Based Memory Baselines2026.06 | 22.4 | 33.68 | |
| Flat Memory ChunkModel=Qwen3-32B, Memory Category=Flat-Text-Based Memory Baselines2026.06 | 20.85 | 32.14 | |
| Full ContextModel=LLaMA-3.3-70B-Instruct, Memory Category=Full Context2026.06 | 20.62 | 31.57 | |
| Full ContextModel=Qwen3-32B, Memory Category=Full Context2026.06 | 19.2 | 30.16 |