Long-context Question Answering on LOCOMO locomo_open
62.02F1 ScoreCAST
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CAST2026.01 | 62.02 | 81.21 | |
| Zep2026.01 | 48.78 | 72.89 | |
| HippoRAG2026.01 | 42.32 | 67.78 | |
| LangMem2026.01 | 40.23 | 69.54 | |
| RAG2026.01 | 36.87 | 56.6 | |
| MRAgentModel (Backbone)=Claude2026.06 | 34.67 | 71.57 | |
| MRAgentModel (Backbone)=Gemini2026.06 | 32.51 | 68.75 | |
| Mem02026.01 | 28.7 | 48.87 | |
| Mem0Model (Backbone)=Claude2026.06 | 28.58 | 56.25 | |
| RAGModel (Backbone)=Claude2026.06 | 26.56 | 43.75 | |
| Mem0Model (Backbone)=Gemini2026.06 | 26.24 | 41.66 | |
| RAGModel (Backbone)=Gemini2026.06 | 25.68 | 41.67 | |
| MemoryOSModel (Backbone)=Gemini2026.06 | 23.43 | 41.66 | |
| LangMemModel (Backbone)=Claude2026.06 | 22.66 | 54.71 | |
| A-MemModel (Backbone)=Claude2026.06 | 22.02 | 47.57 | |
| LangMemModel (Backbone)=Gemini2026.06 | 20.49 | 38.54 | |
| MemoryOSModel (Backbone)=Claude2026.06 | 18.29 | 48.95 | |
| A-MemModel (Backbone)=Gemini2026.06 | 12.49 | 33.33 |