IntentOpt: leaderboard
Intent-based network optimization: from an annotated network diagram and a natural-language intent, the model writes Gurobi code for one of 85 problems in 17 categories (routing, flow, facility placement and others); IntentOpt (Ahmed, Zhu and Choudhury, 2026). Exact match rate in percent, the share of instances whose code runs and reaches an objective within 1% of the verified optimum, direct prompting; higher is better.
Metric: Exact match rate (%; share of 85 network-optimization instances whose generated Gurobi code runs and returns an objective within 1% of the ground truth; direct prompting; annotated network diagram and text intent). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 Mini | 44.71 | #349 |
| 2 | Gemini 2.5 Flash | 41.18 | #452 |
| 3 | Claude Haiku 4.5 | 29.41 | #568 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=intentopt · How It Works · Data refreshed daily, snapshot 2026-11-12.