Coding on LiveBench
40.23AccuracyRAM+
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RAM+Model Category=Merged Models2026.01 | 40.23 | 52.57 | |
| RAM+2026.01 | 40.23 | 52.57 | |
| MemAgent (Memory)Model Category=Base and Task Models2026.01 | 39.25 | 50.12 | |
| TIESModel Category=Merged Models2026.01 | 39.25 | 49.88 | |
| RAMModel Category=Merged Models2026.01 | 38.28 | 49.71 | |
| RAM2026.01 | 38.28 | 49.71 | |
| TAModel Category=Merged Models2026.01 | 38.09 | 51.62 | |
| WUDI2026.01 | 38.09 | 50.95 | |
| CURE (Coding)Model Category=Base and Task Models2026.01 | 37.7 | 49.27 | |
| DARE+TAModel Category=Merged Models2026.01 | 37.5 | 48.6 | |
| DARE+TA2026.01 | 37.5 | 48.6 | |
| FisherModel Category=Merged Models2026.01 | 36.72 | 48.73 | |
| DARE+TIESModel Category=Merged Models2026.01 | 35.93 | 45.66 | |
| ToolRL (Tool)Model Category=Base and Task Models2026.01 | 31.84 | 41.36 | |
| BaseModel Category=Base and Task Models2026.01 | 28.35 | 40.87 |