Janus-1.3B: benchmark results
Provider: DeepSeek. Released 2024-10-17. Access: Open.
Aggregate Rating 1259 ± 40, rank #9173 of 10916 rated models, from 448 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - Gaze Estimation | 45 | Score (%) | 100 |
| OpenVLM MMT-Bench - Musical Instrument Recognition | 100 | Score (%) | 98.3 |
| OpenVLM MMT-Bench - Color Assimilation | 80 | Score (%) | 98.1 |
| OpenVLM MMT-Bench - Disease Diagnosis | 95 | Score (%) | 96.4 |
| OpenVLM MMT-Bench - Counting By Category | 42.5 | Score (%) | 93.7 |
| OpenVLM MMMU - Diagnostics and Laboratory Medicine | 50 | Accuracy (%) | 89.6 |
| OpenVLM MMT-Bench - Small Object Detection | 55 | Score (%) | 88.6 |
| OpenVLM MMT-Bench - Spot The Diff | 80 | Score (%) | 88.6 |
| OpenVLM MMT-Bench - Scene Recognition | 80 | Score (%) | 88.1 |
| OpenVLM MMT-Bench - MIA | 67.5 | Score (%) | 85.7 |
| OpenVLM MMT-Bench - Pixel Localization | 40 | Score (%) | 84.5 |
| OpenVLM MMT-Bench - Profession Recognition | 95 | Score (%) | 84.5 |
Interactive version: theaggregate.ai/model?slug=janus-1-3b · How It Works · Data refreshed daily, snapshot 2026-11-08.