OptiVerse - Stochastic Optimization: leaderboard
Metric: Accuracy (%) on the 120 stochastic optimization problems of OptiVerse; the model writes and runs solver code (gurobi, pyomo, cvxpy, ortools and others) and a problem counts as solved only when every required variable and the objective match the ground truth within 0.1% relative error, extracted and checked by a DeepSeek-V3.2-Chat judge; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 56.67 |
| 2 | Gemini 3 Flash | 54.17 |
| 3 | Gemini 2.5 Pro | 50 |
| 4 | GPT-5.2 (Thinking) | 50 |
| 5 | Gemini 2.5 Flash | 48.33 |
| 6 | O4 Mini | 48.33 |
| 7 | Claude Sonnet 4.5 (Thinking) | 46.67 |
| 8 | O3 | 45.83 |
| 9 | DeepSeek V3.2 (Thinking) | 45 |
| 10 | DeepSeek V3.2 (Non-reasoning) | 45 |
| 11 | Qwen 3 8B (Thinking) | 41.67 |
| 12 | Kimi K2 | 38.33 |
| 13 | Qwen 3 Coder 30B A3B Instruct | 23.33 |
| 14 | Qwen 3 8B (Non-reasoning) | 20.83 |
| 15 | InternLM3-8B-Instruct | 5 |
Interactive version: theaggregate.ai/benchmark?slug=optiverse-stochastic-optimization · How It Works · Data refreshed daily, snapshot 2026-10-07.