Stepfun-ai Models
Explore the Stepfun-ai language and embedding models available through our OpenAI Assistants API-compatible service.
StepFun: Step3
- Context Length:
- 65,536 tokens
- Architecture:
- text+image->text
- Max Output:
- 65,536 tokens
Pricing:
Prompt: $0.00000057
Completion: $0.00000142
Step3 is a 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.
Ready to build with Stepfun-ai?
Start using these powerful models in your applications with our flexible pricing plans.