Optimization Modeling on LogiOR
70.65SAOptiAgent
Evaluation Results
| Method | Links | |
|---|---|---|
| OptiAgentBackbone=GPT 5.42026.07 | 70.65 | |
| OptiAgentBackbone=Sonnet 4.52026.07 | 68.48 | |
| GPT 5.4System Prompt=without2026.07 | 60.87 | |
| GPT 5.4System Prompt=with2026.07 | 54.35 | |
| GPT 5.2System Prompt=with2026.07 | 52.17 | |
| Sonnet 4.5System Prompt=with2026.07 | 51.09 | |
| ORThought2026.07 | 46.01 | |
| SonnetSystem Prompt=without2026.07 | 44.57 | |
| GPT 5.2System Prompt=without2026.07 | 42.39 | |
| CoE2026.07 | 34.78 | |
| DeepSeek-R1System Prompt=with2026.07 | 26.09 | |
| DeepSeekSystem Prompt=without2026.07 | 19.57 | |
| LLMOPT2026.07 | 18.48 | |
| ORLM2026.07 | 15.58 |