Mathematical Optimization Solving on ComplexOR
83.33AccuracyDCM-Agent
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DCM-AgentBackbone=GPT5.12026.04 | 83.33 | — | |
| DCM-AgentBackbone=Deepseek-V3.22026.04 | 77.77 | — | |
| AF-MCTSBackbone=GPT5.12026.04 | 77.77 | — | |
| OptiTreeBackbone=GPT5.12026.04 | 77.77 | — | |
| DCM-AgentBackbone=Qwen3-235B2026.04 | 72.22 | — | |
| OptiTreeBackbone=Deepseek-V3.22026.04 | 72.22 | — | |
| OptiMUSBackbone=GPT5.12026.04 | 72.22 | — | |
| OptiMUSBackbone=Qwen3-235B2026.04 | 66.67 | — | |
| OptiTreeBackbone=Qwen3-235B2026.04 | 66.67 | — | |
| OptiMUSBackbone=Deepseek-V3.22026.04 | 66.67 | — | |
| AF-MCTSBackbone=Deepseek-V3.22026.04 | 66.67 | — | |
| BaselineBackbone=GPT5.12026.04 | 66.67 | — | |
| DCM-AgentBackbone=Qwen3-30B2026.04 | 61.11 | — | |
| AF-MCTSBackbone=Qwen3-235B2026.04 | 61.11 | — | |
| BaselineBackbone=Deepseek-V3.22026.04 | 61.11 | — | |
| OptiTreeBackbone=Qwen3-30B2026.04 | 55.56 | — | |
| BaselineBackbone=Qwen3-235B2026.04 | 55.56 | — | |
| DCM-AgentBackbone=Qwen3-8B2026.04 | 50 | — | |
| OptiMUSBackbone=Qwen3-30B2026.04 | 50 | — | |
| AF-MCTSBackbone=Qwen3-30B2026.04 | 50 | — | |
| OptiTreeBackbone=Qwen3-8B2026.04 | 38.89 | — | |
| BaselineBackbone=Qwen3-30B2026.04 | 38.89 | — | |
| OptiMUSBackbone=Qwen3-8B2026.04 | 33.33 | — | |
| AF-MCTSBackbone=Qwen3-8B2026.04 | 33.33 | — | |
| BaselineBackbone=Qwen3-8B2026.04 | 22.22 | — | |
| CoEParams=Closed, Category=Agentic Methods2026.06 | — | 57.1 | |
| CoTParams=Closed, Category=Agentic Methods2026.06 | — | 39.2 | |
| DeepSeek-V3Params=671B, Category=Zero-shot LLMs2026.06 | — | 44.4 | |
| EvoOptiGraphParams=8B, Category=Fine-tuned LLMs2026.06 | — | 50 | |
| GPT-4oParams=Closed, Category=Zero-shot LLMs2026.06 | — | 38.9 | |
| LLMOPT (origin)Params=14B, Category=Fine-tuned LLMs2026.06 | — | 35.3 | |
| OptiMUS-v0.3Params=Closed, Category=Agentic Methods2026.06 | — | 52.2 | |
| ORLMParams=8B, Category=Fine-tuned LLMs2026.06 | — | 50 | |
| Qwen2.5-72B-InstParams=72B, Category=Zero-shot LLMs2026.06 | — | 38.9 | |
| Qwen3-32BParams=32B, Category=Zero-shot LLMs2026.06 | — | 38.9 |