Open-domain Question Answering on LoCoMo
0.4013F1ShardMemo
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| ShardMemoModel=GPT-OSS-120B2026.01 | 0.4013 | — | — | 0.3382 | — | |
| MEMIRBackbone=GPT-4.12026.05 | 0.342 | — | 0.587 | 0.282 | — | |
| GAMModel=GPT-OSS-120B2026.01 | 0.341 | — | — | 0.2685 | — | |
| LangMemBackbone=GPT-4.12026.05 | 0.3338 | — | — | 0.2726 | — | |
| MEMIRBackbone=GPT-4.1-mini2026.05 | 0.331 | — | 0.619 | 0.281 | — | |
| LangMem*2026.02 | 0.328 | 0.264 | 0.59 | — | — | |
| LangMemBackbone=GPT-4.1-mini2026.05 | 0.328 | — | 0.59 | 0.264 | — | |
| MEMORARetriever Type=Semantic Retriever2026.02 | 0.318 | 0.251 | 0.594 | — | — | |
| NEMORIBackbone=GPT-4.1-mini2026.05 | 0.317 | — | 0.563 | 0.251 | — | |
| MEMORARetriever Type=Policy Retriever2026.02 | 0.308 | 0.246 | 0.594 | — | — | |
| MemoryOSBackbone=GPT-4.12026.05 | 0.307 | — | 0.546 | 0.26 | — | |
| MemboxModel=GPT-4o2026.01 | 0.3061 | 0.2258 | — | — | — | |
| O-MemBackbone=GPT-4.12026.05 | 0.3058 | — | — | 0.2569 | — | |
| Memory-R2(OSS)Training regime=Trained, Answer Agent=GPT-OSS-120B2026.05 | 0.3022 | — | — | 0.245 | 0.7414 | |
| MemoryOSBackbone=GPT-4.1-mini2026.05 | 0.302 | — | 0.604 | 0.256 | — | |
| NEMORIBackbone=GPT-4.12026.05 | 0.302 | — | 0.519 | 0.205 | — | |
| HeLa-MemModel=GPT-4o-mini, Token Length=1,0102026.04 | 0.297 | — | — | 0.2345 | — | |
| MemAgentTraining regime=Trained2026.05 | 0.292 | — | — | 0.2548 | 0.6207 | |
| Mem0Model=GPT-4o-mini2026.01 | 0.2864 | 0.2158 | — | — | — | |
| MEM0Model=GPT-OSS-120B2026.01 | 0.2832 | — | — | 0.2068 | — | |
| SimpleMemBackbone=GPT-4.12026.05 | 0.283 | — | 0.552 | 0.226 | — | |
| LightMEM+AMAModel=GPT-4o2026.01 | 0.2808 | — | — | 0.2236 | — | |
| MemboxModel=GPT-4o-mini2026.01 | 0.2796 | 0.2015 | — | — | — | |
| RAG2026.02 | 0.277 | 0.224 | 0.458 | — | — | |
| SimpleMemBackbone=GPT-4.1-mini2026.05 | 0.269 | — | 0.573 | 0.213 | — | |
| LightMEMModel=GPT-4o2026.01 | 0.2619 | — | — | 0.1995 | — | |
| Nemori*2026.02 | 0.258 | 0.193 | 0.51 | — | — | |
| SwiftMemBackbone=GPT-4.1-mini2026.05 | 0.258 | — | 0.552 | 0.299 | — | |
| Full Context2026.02 | 0.25 | 0.204 | 0.5 | — | — | |
| HeLa-MemModel=GPT-4o, Token Length=1,0362026.04 | 0.2438 | — | — | 0.1924 | — | |
| LightMEM+AMAModel=GPT-4o-mini2026.01 | 0.2422 | — | — | 0.1817 | — | |
| HeLa-MemModel=Qwen2.5-14b, Token Length=9442026.04 | 0.2422 | — | — | 0.2023 | — | |
| Zep*2026.02 | 0.242 | 0.193 | 0.438 | — | — | |
| ZepBackbone=GPT-4.1-mini2026.05 | 0.242 | — | 0.438 | 0.193 | — | |
| Nemori+AMAModel=GPT-4o-mini2026.01 | 0.2415 | — | — | 0.1735 | — | |
| Nemori+AMAModel=GPT-4o2026.01 | 0.2405 | — | — | 0.181 | — | |
| LightMemBackbone=GPT-4.12026.05 | 0.238 | — | 0.502 | 0.199 | — | |
| MemoryOSModel=GPT-4o-mini, Token Length=2,0002026.04 | 0.2375 | — | — | 0.1717 | — | |
| Mem0Backbone=GPT-4.1-mini2026.05 | 0.237 | — | 0.479 | 0.177 | — | |
| Memory-R1Training regime=Trained2026.05 | 0.2355 | — | — | 0.2071 | 0.4096 | |
| NemoriModel=GPT-4o-mini2026.01 | 0.2348 | — | — | 0.1778 | — | |
| Mem0Backbone=GPT-4.12026.05 | 0.234 | — | 0.451 | 0.174 | — | |
| NemoriModel=GPT-4o2026.01 | 0.2322 | — | — | 0.1728 | — | |
| Mem0Model=GPT-4o, Local re-implementation=true, k=302026.01 | 0.2304 | 0.1762 | — | — | — | |
| LightMEMModel=GPT-4o-mini2026.01 | 0.2283 | — | — | 0.1696 | — | |
| MEM1Training regime=Trained2026.05 | 0.2267 | — | — | 0.1472 | 0.431 | |
| Mem0Model=GPT-4o-mini, Local re-implementation=true, k=302026.01 | 0.2257 | 0.1654 | — | — | — | |
| HIMEMBackbone=GPT-4.1-mini2026.05 | 0.219 | — | 0.434 | 0.102 | — | |
| RAGModel=GPT-OSS-120B2026.01 | 0.2154 | — | — | 0.1529 | — | |
| MemoryOSTraining regime=Training-free2026.05 | 0.2103 | — | — | 0.1841 | 0.3855 | |
| Memory-R2Training regime=Trained2026.05 | 0.2076 | — | — | 0.1678 | 0.6753 | |
| Mem02026.02 | 0.206 | 0.153 | 0.5 | — | — | |
| HIMEMBackbone=GPT-4.12026.05 | 0.206 | — | 0.382 | 0.09 | — | |
| MemoryOSModel=GPT-4o, Token Length=2,0002026.04 | 0.2058 | — | — | 0.1585 | — | |
| Mem0Training regime=Training-free2026.05 | 0.2031 | — | — | 0.1582 | 0.4819 | |
| MemoryOSModel=Qwen2.5-14b2026.04 | 0.2027 | — | — | 0.1594 | — | |
| MEMORYOSModel=GPT-OSS-120B2026.01 | 0.1983 | — | — | 0.144 | — | |
| LightMemBackbone=GPT-4.1-mini2026.05 | 0.192 | — | 0.524 | 0.177 | — | |
| LoCoMoBackbone=GPT-4.1-mini2026.05 | 0.191 | — | — | 0.171 | — | |
| A-MEMModel=GPT-4o, Local re-implementation=true, k=302026.01 | 0.1745 | 0.1558 | — | — | — | |
| LIGHTMEMModel=GPT-OSS-120B2026.01 | 0.1713 | — | — | 0.1381 | — | |
| A-MEMModel=GPT-4o2026.01 | 0.171 | 0.1584 | — | — | — | |
| A-MemModel=GPT-4o, Token Length=1,2162026.04 | 0.171 | — | — | 0.1584 | — | |
| A-MEMModel=GPT-4o-mini, Local re-implementation=true, k=302026.01 | 0.166 | 0.138 | — | — | — | |
| A-MemModel=Qwen2.5-3b, Token Length=1,3002026.04 | 0.1648 | — | — | 0.1431 | — | |
| LoCoMoModel=GPT-4o2026.01 | 0.1647 | — | — | 0.148 | — | |
| LoCoMoModel=GPT-4o2026.01 | 0.1647 | 0.148 | — | — | — | |
| LoCoMoModel=GPT-4o, Token Length=16,9102026.04 | 0.1647 | — | — | 0.148 | — | |
| LightMemModel=Qwen2.5-14b2026.04 | 0.1581 | — | — | 0.1181 | — | |
| MEM0Model=Qwen2.5-14b2026.04 | 0.1503 | — | — | 0.1128 | — | |
| A-MEMTraining regime=Training-free2026.05 | 0.1473 | — | — | 0.1266 | 0.3132 | |
| A-MEMModel=GPT-4o2026.01 | 0.1463 | — | — | 0.1717 | — | |
| Vanilla LLMModel=GPT-OSS-120B2026.01 | 0.1425 | — | — | 0.1186 | — | |
| A-MemBackbone=GPT-4.12026.05 | 0.142 | — | 0.405 | 0.154 | — | |
| A-MEM+AMAModel=GPT-4o2026.01 | 0.1417 | — | — | 0.1323 | — | |
| MemoryOSModel=Qwen2.5-3b, Token Length=2,0002026.04 | 0.1357 | — | — | 0.1086 | — | |
| A-MemModel=Qwen2.5-14b, Token Length=1,3002026.04 | 0.1349 | — | — | 0.1074 | — | |
| A-MemBackbone=GPT-4.1-mini2026.05 | 0.134 | — | 0.375 | 0.127 | — | |
| MemGPTModel=GPT-4o2026.01 | 0.1224 | — | — | 0.1187 | — | |
| MEMGPTModel=GPT-4o2026.01 | 0.1224 | 0.1187 | — | — | — | |
| MemGPTModel=GPT-4o, Token Length=16,9872026.04 | 0.1224 | — | — | 0.1187 | — | |
| HeLa-MemModel=Qwen2.5-3b, Token Length=1,0722026.04 | 0.1224 | — | — | 0.1024 | — | |
| A-MEMModel=GPT-4o-mini2026.01 | 0.1214 | 0.12 | — | — | — | |
| A-MemModel=GPT-4o-mini, Token Length=2,5202026.04 | 0.1214 | — | — | 0.12 | — | |
| LoCoMoModel=GPT-4o-mini2026.01 | 0.1204 | — | — | 0.1116 | — | |
| LoCoMoModel=GPT-4o-mini2026.01 | 0.1204 | 0.1116 | — | — | — | |
| LoCoMoModel=GPT-4o-mini, Token Length=16,9102026.04 | 0.1204 | — | — | 0.1116 | — | |
| A-MEMModel=GPT-OSS-120B2026.01 | 0.1186 | — | — | 0.111 | — | |
| RAGTraining regime=Training-free2026.05 | 0.1184 | — | — | 0.1002 | 0.1928 | |
| A-MEMModel=GPT-4o-mini2026.01 | 0.103 | — | — | 0.093 | — | |
| MemGPTModel=GPT-4o-mini2026.01 | 0.0915 | — | — | 0.0744 | — | |
| MEMGPTModel=GPT-4o-mini2026.01 | 0.0915 | 0.0744 | — | — | — | |
| MemGPTModel=GPT-4o-mini, Token Length=16,9772026.04 | 0.0915 | — | — | 0.0744 | — | |
| ReadAgentModel=GPT-4o2026.01 | 0.0884 | — | — | 0.0837 | — | |
| READAGENTModel=GPT-4o2026.01 | 0.0884 | 0.0837 | — | — | — | |
| ReadAgentModel=GPT-4o, Token Length=8052026.04 | 0.0884 | — | — | 0.0837 | — | |
| A-MEM+AMAModel=GPT-4o-mini2026.01 | 0.087 | — | — | 0.083 | — | |
| ReadAgentBackbone=GPT-4.1-mini2026.05 | 0.077 | — | — | 0.066 | — | |
| MemGPTModel=Qwen2.5-3b, Token Length=16,9612026.04 | 0.0704 | — | — | 0.071 | — | |
| MemoryBankModel=Qwen2.5-3b, Token Length=2982026.04 | 0.0663 | — | — | 0.0658 | — |