ProLLM - Image Understanding: leaderboard
ProLLM benchmark measuring LLM image comprehension and caption quality across business contexts.
Metric: Score (%). Source: www.prollm.ai. Saturation forecast: Estimated already saturated. 27 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 | 89.3 | #124 |
| 2 | O1 | 88.6 | #183 |
| 3 | GPT-4o | 87.9 | #343 |
| 4 | GPT-4.1 | 83.9 | #248 |
| 5 | GPT-4.5 | 83.2 | #274 |
| 6 | GPT-4.1 Mini | 81.2 | #355 |
| 7 | Gemini 2.5 Pro | 80.5 | #152 |
| 8 | O4 Mini | 75.8 | #180 |
| 9 | GPT-4o Mini | 73.2 | #599 |
| 10 | Gemini 2.0 Flash | 71.1 | #342 |
| 11 | Llama 4 Scout Instruct | 61.7 | #539 |
| 12 | Llama 4 Maverick Instruct | 58.4 | #456 |
| 13 | GPT-4.1 Nano | 57 | #731 |
| 14 | Qwen 2.5 VL 72B Instruct | 56.4 | #377 |
| 15 | Qwen 2.5 VL 7B Instruct | 52.3 | #654 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=prollm-image-understanding · How It Works · Data refreshed daily, snapshot 2026-10-20.