GPT-4v (0409, detail-low): benchmark results
Provider: OpenAI. Released 2024-04-09. Access: API.
Aggregate Rating 1537 ± 34, rank #1861 of 10916 rated models, from 201 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM SEEDBench IMG - Instance Interaction | 81.4 | Accuracy (%) | 98.4 |
| OpenVLM MMBench V1.1 CN - Celebrity Recognition | 93.9 | Accuracy (%) | 96.3 |
| OpenVLM MMMU - Art Theory | 93.3 | Accuracy (%) | 96.1 |
| OpenVLM MMBench V1.1 EN - Celebrity Recognition | 93.9 | Accuracy (%) | 95.2 |
| OpenVLM MME - Posters | 187.8 | Score | 94 |
| OpenVLM MMMU - Basic Medical Science | 76.7 | Accuracy (%) | 93.6 |
| OpenVLM MMMU - Public Health | 90 | Accuracy (%) | 93.6 |
| OpenVLM MMBench V1.1 EN - OCR | 98.9 | Accuracy (%) | 93.3 |
| OpenVLM LLaVA-Bench - Complex | 103.8 | Score | 93.1 |
| OpenVLM MMMU - Art | 80 | Accuracy (%) | 92.8 |
| OpenVLM MMMU - Health and Medicine | 72 | Accuracy (%) | 92 |
| OpenVLM MME - Count | 175 | Score | 91.5 |
Interactive version: theaggregate.ai/model?slug=gpt-4v-0409-detail-low · How It Works · Data refreshed daily, snapshot 2026-11-08.