IDEFICS2-8B: benchmark results
Provider: HuggingFace. Released 2024-04-15. Access: Open.
Aggregate Rating 1425 ± 20, rank #4906 of 10916 rated models, from 620 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - Multiple Image Captioning | 100 | Score (%) | 97.6 |
| OpenVLM MMT-Bench - Spot The Diff | 95 | Score (%) | 97.3 |
| HELM VHELM - Blink Perception - Relative Reflectance | 68.66 | Prefix Quasi-Exact Match (%) | 95.7 |
| HELM VHELM - Vqa Base | 87 | PEM | 95.7 |
| HELM VHELM - Vqa Dialect | 84.1 | PEM (Fairness) | 95.7 |
| OpenVLM MMT-Bench - Human Object Interaction Recognition | 90 | Score (%) | 95.6 |
| HELM VHELM - Vqa Robustness | 81.5 | EM (Robustness) | 95.2 |
| OpenVLM MMBench V1.1 EN - Image Topic | 98.9 | Accuracy (%) | 95.2 |
| OpenVLM MMT-Bench - Facial Expression Recognition | 75 | Score (%) | 94.4 |
| OpenVLM MMT-Bench - Logo and Brand Recognition | 100 | Score (%) | 94.4 |
| OpenVLM MMT-Bench - Celebrity Recognition | 95 | Score (%) | 94.2 |
| OpenVLM MMT-Bench - Color Constancy | 71.4 | Score (%) | 94.2 |
Interactive version: theaggregate.ai/model?slug=idefics2-8b · How It Works · Data refreshed daily, snapshot 2026-11-08.