Models

Open reasoning model trained with large-scale reinforcement learning. Excels at math, code and multi-step logic.

deepseek/deepseek-r1128K context$1.35/M input$5.4/M output

Hybrid reasoning model with sparse attention for long contexts. Strong at coding, tool use and agentic workflows.

deepseek/deepseek-v3.2128K context$0.62/M input$1.85/M output