Multi-Agent System Performance on Research
77.62TS ScoreLLMA-Mem
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| LLMA-MemModel=Claude-Sonnet-4.52026.03 | 77.62 | 74.11 | 77.77 | |
| MARBLEModel=Deepseek-v3.22026.03 | 76.3 | 58.21 | 68.43 | |
| A-MemModel=Claude-Sonnet-4.52026.03 | 75.69 | 82.47 | 77.79 | |
| W.o. MemoryModel=Deepseek-v3.22026.03 | 75.53 | 58.46 | 65.94 | |
| MARBLEModel=Claude-Sonnet-4.52026.03 | 75.17 | 81.84 | 78.37 | |
| W.o. MemoryModel=Claude-Sonnet-4.52026.03 | 75.12 | 78.97 | 76.94 | |
| LLMA-MemModel=Deepseek-v3.22026.03 | 75.1 | 65.63 | 71.86 | |
| A-MemModel=Qwen3-32B-Instruct2026.03 | 73.33 | 68.61 | 70.82 | |
| LLMA-MemModel=Qwen3-32B-Instruct2026.03 | 73.17 | 73.08 | 73.5 | |
| A-MemModel=Deepseek-v3.22026.03 | 72.4 | 55.64 | 64.27 | |
| MARBLEModel=Qwen3-32B-Instruct2026.03 | 72.38 | 68.62 | 70.36 | |
| W.o. MemoryModel=Qwen3-32B-Instruct2026.03 | 71.56 | 72.63 | 72.06 | |
| A-MemModel=Qwen3-next-80B2026.03 | 69.1 | 48.39 | 57.67 | |
| LLMA-MemModel=Qwen3-next-80B2026.03 | 67.89 | 43.62 | 49.54 | |
| MARBLEModel=Qwen3-next-80B2026.03 | 61.62 | 42.66 | 55.11 | |
| W.o. MemoryModel=Qwen3-next-80B2026.03 | 57.22 | 45.52 | 50.13 |