Step-1o: benchmark results
Provider: StepFun. Access: API.
Aggregate Rating 1622 ± 19, rank #818 of 10916 rated models, from 526 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLaVA-Bench | 107.3 | Score | 100 |
| OpenVLM CCBench - Scenery and Buildings | 98.9 | Accuracy (%) | 100 |
| OpenVLM LLaVA-Bench - Conv | 107.5 | Score | 100 |
| OpenVLM MMBench V1.1 CN - Celebrity Recognition | 98.6 | Accuracy (%) | 100 |
| OpenVLM MMBench V1.1 CN - Future Prediction | 89.3 | Accuracy (%) | 100 |
| OpenVLM MMBench V1.1 EN - Celebrity Recognition | 98.6 | Accuracy (%) | 100 |
| OpenVLM MMMU - Art | 96.7 | Accuracy (%) | 100 |
| OpenVLM MMMU - Biology | 80 | Accuracy (%) | 100 |
| OpenVLM MMMU - Clinical Medicine | 83.3 | Accuracy (%) | 100 |
| OpenVLM MMMU - Design | 93.3 | Accuracy (%) | 100 |
| OpenVLM MMT-Bench - AND | 73 | Score (%) | 100 |
| OpenVLM MMT-Bench - Face Mask Anomaly Detection | 80 | Score (%) | 100 |
Interactive version: theaggregate.ai/model?slug=step-1o · How It Works · Data refreshed daily, snapshot 2026-11-08.