OptimAI: Operations Research Assistant

A Qwen2.5-Math-7B fine-tune (SFT + DPO) for OR problems: LP, IP, shortest path, queueing, KKT, duality, etc. Try an example or paste your own.

Problem

Max new tokens

128 2048

Model output

Examples

Model: billwang37/WWang-Lab-OptimAI
Note: First call takes ~30-60s (cold start). Subsequent calls ~10-20s.