StructEval: leaderboard
Structured-output benchmark evaluating text and visual structured generation and conversion across 18 formats and 2,035 examples.
Metric: Average (imported). Source: github.com. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 76.02 | #343 |
| 2 | GPT-4.1 Mini | 75.64 | #355 |
| 3 | O1 Mini | 75.58 | #323 |
| 4 | GPT-4o Mini | 73.19 | #599 |
| 5 | Gemini 1.5 Pro | 71.75 | #445 |
| 6 | Qwen 3 4B | 67.04 | #829 |
| 7 | Gemini 2.0 Flash | 62.55 | #342 |
| 8 | Llama 3.1 8B Instruct | 61.77 | #1035 |
| 9 | Qwen 2.5 7B Instruct | 59.03 | #862 |
| 10 | Phi-4 Mini Instruct | 56.97 | #1245 |
| 11 | Llama 3 8B Instruct | 51.59 | #1129 |
| 12 | Phi-3 Mini 128K Instruct | 40.79 | #1298 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=structeval · How It Works · Data refreshed daily, snapshot 2026-10-20.