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.
Added Jul 31, 2025
Context Window
65.5K
Max Output
8.2K
Input Price (Auto)
$0.25/1M
Output Price (Auto)
$0.65/1M
Performance metrics and benchmarks
Sourced from Artificial Analysis.
No benchmark data is available yet for this model.
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
