GPT-4o Mini (0718, detail-high): benchmark results
Provider: OpenAI. Released 2024-07-18. Access: API.
Aggregate Rating 1547 ± 24, rank #1733 of 10916 rated models, from 416 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - Rock Recognition | 85 | Score (%) | 100 |
| OpenVLM MMT-Bench - TU | 57.5 | Score (%) | 99.8 |
| OpenVLM MMT-Bench - Next Image Prediction | 90 | Score (%) | 99.5 |
| OpenVLM MMT-Bench - Anatomy Identification | 90 | Score (%) | 99.3 |
| OpenVLM MME - OCR | 200 | Score | 99.1 |
| OpenVLM MMBench V1.1 CN - OCR | 100 | Accuracy (%) | 98.6 |
| OpenVLM MMMU - Literature | 93.3 | Accuracy (%) | 98.1 |
| OpenVLM MMT-Bench - Animated Character Recognition | 85 | Score (%) | 98.1 |
| OpenVLM MMT-Bench - MeViS | 55 | Score (%) | 98.1 |
| OpenVLM MMT-Bench - Color Recognition | 90 | Score (%) | 97.8 |
| OpenVLM MMT-Bench - Facial Expression Recognition | 80 | Score (%) | 97.3 |
| OpenVLM MMT-Bench - Abstract Visual Recognition | 90 | Score (%) | 96.8 |
Interactive version: theaggregate.ai/model?slug=gpt-4o-mini-0718-detail-high · How It Works · Data refreshed daily, snapshot 2026-11-08.