Step3 is our cutting-edge multimodal reasoning model—built on a Mixture-of-Experts architecture with 321B total parameters and 38B active. It is designed end-to-end to minimize decoding costs while delivering top-tier performance in vision–language reasoning. Through the co-design of Multi-Matrix Factorization Attention (MFA) and Attention-FFN Disaggregation (AFD), Step3 maintains exceptional efficiency across both flagship and low-end accelerators.

You can access Step3's API on https://platform.stepfun.com/ , we provide OpenAI/Anthropic-compatible API for you.

Our model checkpoints are stored in bf16 and block-fp8 format, you can find it on Huggingface.

Currently, it is recommended to run Step3 on the following inference engines:

• vLLM
• SGLang

Deployment and Request examples for vLLM and SGLang can be found in the Model Deployment Guide.

If you have any questions, please reach out at [email protected] .

Both the code repository and the model weights are released under the Apache License (Version 2.0).

@misc{step3system,
      title={Step-3 is Large yet Affordable: Model-system Co-design for Cost-effective Decoding}, 
      author={StepFun Team},
      year={2025},
      eprint={2507.19427},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2507.19427}, 
}

@misc{step3blog,
      title={Step3: Cost-Effective Multimodal Intelligence}, 
      author={StepFun Team},
      url={https://stepfun.ai/research/step3}, 
}