Creator
StepFun
Released
2026-01-20
Intelligence
9.5
Artificial Analysis Index
Coding
—
Artificial Analysis Index
In $/1M
$0.00
input tokens
Out $/1M
$0.00
output tokens
Blended $/1M
$0.00
3:1 blended
Speed
0
tokens / sec
Profile
License, openness, and modality — the governance layer.
License
Apache 2.0
commercial OK
Weights
Open
downloadable
Modalities
—
Parameters
10B
total
STEP3-VL-10B is a lightweight open-source foundation model designed to redefine the trade-off between compact efficiency and frontier-level multimodal intelligence. Built on a unified, fully unfrozen pre-training strategy on 1.2T multimodal tokens integrating a language-aligned Perception Encoder with a Qwen3-8B decoder. Features Parallel Coordinated Reasoning (PaCoRe) to scale test-time compute for complex perceptual reasoning.
Capability profile
Category strength across 20 domains, via LLM Stats.
Benchmark breakdown
Independent evaluation scores, normalized to 0–100.