OptiVerse - Game Optimization: leaderboard

OptiVerse (2026) is a benchmark of textbook-sourced optimization problems that models solve by writing and running solver code, a problem counting as solved only when every required value matches the ground truth within 0.1%. This board is accuracy in percent on the game optimization problems; higher is better.

Metric: Accuracy (%) on the 51 game 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 February 2027. 20 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Gemini 3 Flash56.86#180
2Gemini 2.5 Flash54.9#466
3GPT-5.2Thinking54.9#190
4Gemini 3 Pro50.98#146
5DeepSeek V3.2Thinking49.02#425
6Qwen3-235B-Thinking (OptiVerse checkpoint unspecified)49.02
7O4 Mini47.06#372
8Claude Sonnet 4.5Thinking45.1#267
9O345.1#242
10Gemini 2.5 Pro43.14#280
11Qwen3-30B-Thinking (OptiVerse checkpoint unspecified)43.14
12Qwen 3 235B Instruct (OptiVerse checkpoint unspecified)41.18
13Qwen 3 8BThinking37.25#1890
14GPT-OSS-120B35.29#712
15Kimi K225.49#448

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=optiverse-game-optimization · How It Works · Data refreshed daily, snapshot 2026-11-22.