StepFun · Step 3
Step3 VL 10B
Small open-weight StepFun vision-language reasoning model for local/edge multimodal use.
- AA Intelligence
- 7.7#115 of 142
- Price
- Self-host
- Output speed
- —
- Context window
- 66K tokens
Composite indexes
- Artificial Analysis Intelligence Index#115 of 1428
Leader: Claude Opus 5.5 at 58
Agents and tool use
- τ²-Bench#70 of 8816.1%
Leader: GLM-5.2 at 99.1%
Reasoning and knowledge
- Humanity’s Last Exam#100 of 14710.8%
Leader: Claude Opus 5.5 at 61.4%
- GPQA Diamond#99 of 14269%
Leader: GPT-6 Astra at 96.1%
- CritPt#98 of 1390%
Leader: GPT-5.6 Sol at 32.3%
Vision and long context
- AA Long Context Reasoning#128 of 1390%
Leader: Kimi K3 at 88.7%
- MMMU-Pro#53 of 7264%
Leader: Claude Opus 5.5 at 87.7%
Other published results
| Result | Score | Reported by | Source |
|---|---|---|---|
| Terminal-Bench Hard (AA) · AA | 5.3% | independent | Artificial Analysis: Step3 VL 10B ↗ |
| IFBench (AA) · AA | 50.2% | independent | Artificial Analysis: Step3 VL 10B ↗ |
| AA-Omniscience Index · AA | -59 index (-100..100) | independent | Artificial Analysis: Step3 VL 10B ↗ |