Natural Language to Optimization on NL4Opt (test)
94.5Success RateNC2C
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| NC2CModel=GPT-5.12026.01 | 94.5 | 96.7 | |
| NC2CModel=Qwen3-235B-A22B2026.01 | 91.2 | 94.4 | |
| OptiMUSModel=GPT-5.12026.01 | 90.2 | 94 | |
| Chain-of-ExpertsModel=GPT-5.12026.01 | 88.3 | 92.5 | |
| ReflexionModel=GPT-5.12026.01 | 87.5 | 92.8 | |
| VanillaModel=GPT-5.12026.01 | 85 | 90.2 | |
| OptiMUSModel=Qwen3-235B-A22B2026.01 | 85 | 90.5 | |
| Chain-of-ExpertsModel=Qwen3-235B-A22B2026.01 | 82.3 | 88.5 | |
| ReflexionModel=Qwen3-235B-A22B2026.01 | 80.8 | 88 | |
| VanillaModel=Qwen3-235B-A22B2026.01 | 78.5 | 85.2 |