Step 3
Step 3 is StepFun’s open-weight multimodal reasoning model with a sparse language backbone and visual encoder. Multi-head latent attention reduces serving overhead while image-conditioned reasoning supports documents, charts, and general visual questions. The language backbone has 316B parameters; the full vision-language model has 321B. Apache-licensed weights and a technical report support private deployment and reproducible evaluation. The full weight footprint, context length, and selected reasoning mode affect deployment cost and latency.
2025-07-31
321B total, 38B active
Mixture-of-Experts multimodal Transformer
Apache-2.0
Specifications
- Parameters
- 321B total, 38B active
- Architecture
- Mixture-of-Experts multimodal Transformer
- License
- Apache-2.0
- Context Window
- 65,536 tokens
- Type
- multimodal
- Modalities
- textimage
Benchmark Scores
Advanced Specifications
- Model Family
- Step
- API Access
- Not Available
- Chat Interface
- Not Available
Capabilities & Limitations
- Capabilities
- visual reasoningSTEM reasoninglanguage generation
- Known Limitations
- Sparse models still require storage for all weightsParallel reasoning consumes additional inference computeEvaluation settings strongly affect reported scores
- Notable Use Cases
- private assistantscoding or document researchmultimodal experimentation