ChinaReasoning128K context671B MoE

Explore DeepSeek-R1

Run DeepSeek's reasoning model for mathematics, software engineering, planning, and difficult multi-step analysis.

Model

deepseek-reasoner

Input

Configure your request

Conversation history (JSON) Optional

Reasoner accepts user and assistant conversation turns.

Reasoning task

Parameters

Stream reasoning and answer

Maximum output tokens

Includes reasoning and final-answer tokens.

Response format

Stop sequences (JSON)