Tool Use on LiveMCPBench
77LiveMCP ScoreGEPA
Evaluation Results
| Method | Links | |
|---|---|---|
| GEPAModel=Kimi-k2.62026.07 | 77 | |
| NREModel=gpt-oss-120b2026.07 | 76 | |
| NREModel=Mistral 3.52026.07 | 75 | |
| NREModel=gpt-5.4-mini2026.07 | 75 | |
| VanillaModel=Opus 4.72026.07 | 75 | |
| NREModel=Kimi-k2.62026.07 | 74.7 | |
| VanillaModel=Sonnet 4.62026.07 | 74.7 | |
| NREModel=Deepseek-v32026.07 | 73.7 | |
| VanillaModel=Kimi-k2.62026.07 | 72 | |
| VanillaModel=Haiku 4.52026.07 | 70.5 | |
| NREModel=Haiku 4.52026.07 | 69 | |
| GEPAModel=gpt-oss-120b2026.07 | 63.2 | |
| VanillaModel=Deepseek-v32026.07 | 54 | |
| VanillaModel=gpt-oss-120b2026.07 | 53 | |
| VanillaModel=Deepseek-R12026.07 | 43 | |
| VanillaModel=gpt-5.4-mini2026.07 | 42 | |
| GEPAModel=Deepseek-v32026.07 | 40 | |
| RLMModel=Kimi-k2.62026.07 | 14.7 | |
| RLMModel=gpt-oss-120b2026.07 | 5.3 | |
| RLMModel=Deepseek-v32026.07 | 0 |