IDEFICS-9B-Instruct: benchmark results
Provider: HuggingFace. Released 2023-08-22. Access: Open.
Aggregate Rating 1196 ± 40, rank #9640 of 10916 rated models, from 557 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - Spot The Diff | 100 | Score (%) | 99.5 |
| OpenVLM COCO Captions | 90.4 | CIDEr | 97.6 |
| OpenVLM MMT-Bench - Gaze Estimation | 35 | Score (%) | 93 |
| OpenVLM MMT-Bench - MIA | 72.5 | Score (%) | 91.3 |
| OpenVLM MMT-Bench - Profession Recognition | 95 | Score (%) | 84.5 |
| HELM Image2Struct - Image2Latex Wild | 64.4 | EMS | 84.2 |
| OpenVLM MMT-Bench - Deepfake Detection | 60 | Score (%) | 83 |
| OpenVLM MMT-Bench - GUI Install | 50 | Score (%) | 82.5 |
| OpenVLM MMT-Bench - Color Constancy | 57.1 | Score (%) | 82.3 |
| OpenVLM MMT-Bench - Animal Keypoint Detection | 40 | Score (%) | 82 |
| OpenVLM MMT-Bench - Color Contrast | 65 | Score (%) | 81.8 |
| OpenVLM MMT-Bench - Sign Language Recognition | 40 | Score (%) | 81.3 |
Interactive version: theaggregate.ai/model?slug=idefics-9b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.