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.
Recent activity on Step3
Total usage per day on OpenRouter
Prompt
159K
Reasoning
57K
Completion
29K
Prompt tokens measure input size. Reasoning tokens show internal thinking before a response. Completion tokens reflect total output length.