OR-Space - Explain: leaderboard
Metric: Explain rubric score (0-100): exact coverage, reasoning, grounding and answer quality of short solver-grounded reports minus a hallucination penalty, checklist items scored programmatically or by a judge model, 100 industrial optimization instances in a filesystem workspace (docs, data and an empty or heuristic src directory), Gurobi 12.0.1 track, API temperature at most 0.1; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 | 86.52 |
| 2 | GPT-5 Mini | 82.94 |
| 3 | Claude Opus 4.6 | 80.2 |
| 4 | Claude Sonnet 4.5 (Thinking) | 79.63 |
| 5 | Qwen 3.5 27B | 77.88 |
| 6 | DeepSeek V4 Flash | 77.77 |
| 7 | Claude Sonnet 4.5 | 77.58 |
| 8 | GPT-5.1 | 76.16 |
| 9 | Qwen 3 Max | 74.66 |
| 10 | Gemini 3.1 Pro (Preview) | 73 |
| 11 | DeepSeek V4 Pro | 68.87 |
| 12 | Qwen 3 32B (Non-reasoning) | 61.32 |
| 13 | Qwen 3 32B | 61.28 |
| 14 | Qwen 3 14B (Non-reasoning) | 61.03 |
| 15 | GPT-4o | 59.36 |
Interactive version: theaggregate.ai/benchmark?slug=or-space-explain · How It Works · Data refreshed daily, snapshot 2026-10-07.