← All models

DeepSeek R1

im/together-deepseek-r1

stable

DeepSeek R1 reasoning model. Shows full chain-of-thought, strong math/code.

Specs

Capabilities and limits

Provider
Together AI
Model family
deepseek-r1
Type
reasoning
Context length
131,072 tokens
Max output
32,768 tokens
Streaming
Yes
Tool calling
No
JSON mode
No
Vision
No

Pricing

Active credit rate

Input
3600000 cr / M tokens · $3.6000
Output
8400000 cr / M tokens · $8.4000
Effective from
5/17/2026

Best for

mathcodingstep-by-step-reasoningresearch

Endpoints

Routes this model serves

  • /v1/chat/completions