step3
Provider: SiliconFlow (China), Context: 66000, Output Limit: 66000
| Provider | Source | Input Price ($/1M) | Output Price ($/1M) | Description | Free |
|---|---|---|---|---|---|
| siliconflowcn | models-dev | Input: $0.57 | Output: $1.42 | Provider: SiliconFlow (China), Context: 66000, Output Limit: 66000 | |
| siliconflow | models-dev | Input: $0.57 | Output: $1.42 | Provider: SiliconFlow, Context: 66000, Output Limit: 66000 | |
| openrouter | openrouter | Input: $0.57 | Output: $1.42 | 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. Context: 65536 |