GPT-4v (0409, detail-high): benchmark results
Provider: OpenAI. Released 2024-04-09. Access: API.
Aggregate Rating 1559 ± 33, rank #1566 of 10916 rated models, from 422 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - Fashion Recognition | 95 | Score (%) | 100 |
| OpenVLM MMT-Bench - Interactive Segmentation | 85.7 | Score (%) | 99 |
| OpenVLM MMT-Bench - Salient Object Detection RGBD | 70 | Score (%) | 99 |
| OpenVLM MMMU - Art | 86.7 | Accuracy (%) | 98.2 |
| OpenVLM MMT-Bench - Astronomical Recognition | 100 | Score (%) | 97.8 |
| OpenVLM MMT-Bench - Handwritten Retrieval | 85 | Score (%) | 97.8 |
| OpenVLM MMBench V1.1 EN - Celebrity Recognition | 95.2 | Accuracy (%) | 97.5 |
| OpenVLM Q-Bench - How (In-context Distortion) | 72.4 | Accuracy (%) | 97.5 |
| OpenVLM MMT-Bench - Face Retrieval | 100 | Score (%) | 97.3 |
| OpenVLM MMBench V1.1 CN - Celebrity Recognition | 94.6 | Accuracy (%) | 97.2 |
| OpenVLM LLaVA-Bench - Conv | 100 | Score | 97 |
| OpenVLM MMMU - Public Health | 93.3 | Accuracy (%) | 96.8 |
Interactive version: theaggregate.ai/model?slug=gpt-4v-0409-detail-high · How It Works · Data refreshed daily, snapshot 2026-11-08.