MM-IFEval — leaderboard
Metric: Overall Accuracy (%). Source: huggingface.co. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 (2025-04-14) | 72.81 |
| 2 | Gemini 2.5 Flash | 71.28 |
| 3 | Llama 4 Scout Instruct | 68.53 |
| 4 | Gemini 2.0 Flash | 67.24 |
| 5 | Claude 3.7 Sonnet | 64.38 |
| 6 | InternVL3-78B | 63.85 |
| 7 | Claude 3.5 Sonnet (20241022) | 63.7 |
| 8 | Grok 2 (1212) | 61.52 |
| 9 | InternVL2.5-78B | 58.36 |
| 10 | InternVL3-38B | 57.94 |
| 11 | Pixtral-12B | 55.25 |
| 12 | InternVL3-8B | 51.79 |
| 13 | Llama 3.2 11B Instruct | 46.73 |
Interactive version: aibenchmarks.dev/benchmark?slug=mm-ifeval · How the rankings work · Data refreshed daily, snapshot 2026-07-13.