DeepSeek-VL-7B: benchmark results
Provider: DeepSeek. Released 2024-03-11.
Aggregate Rating 1345 ± 30, rank #7845 of 10916 rated models, from 463 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - Fashion Recognition | 90 | Score (%) | 99 |
| OpenVLM MMT-Bench - One Shot Detection | 95 | Score (%) | 99 |
| OpenVLM A-Bench - Nouns as Adjectives | 86 | Accuracy (%) | 97.2 |
| OpenVLM MMT-Bench - Temporal Sequence Understanding | 55 | Score (%) | 97.1 |
| OpenVLM MMT-Bench - CIM | 75 | Score (%) | 96.4 |
| OpenVLM MMT-Bench - Animals Recognition | 100 | Score (%) | 94.9 |
| OpenVLM MMT-Bench - Logo and Brand Recognition | 100 | Score (%) | 94.4 |
| OpenVLM MMT-Bench - Rock Recognition | 75 | Score (%) | 94.2 |
| OpenVLM MMT-Bench - Depth Estimation | 80 | Score (%) | 93.7 |
| OpenVLM MMT-Bench - Color Assimilation | 75 | Score (%) | 93 |
| FAQ (Deepfake Forensics) - Facial Perception | 54.7 | Accuracy (%) on FAQ's Level 1 facial-perception questions: f | 91.7 |
| OpenVLM OCR-VQA | 67 | Accuracy (%) | 91 |
Interactive version: theaggregate.ai/model?slug=deepseek-vl-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.