DeepSeek-R1 is an open-weights reasoning model developed by DeepSeek, optimized for generating long chains of thought. It is trained with reinforcement learning, making it suitable for complex reasoning tasks involving extended text contexts. Its large context window supports handling lengthy inputs.

What is DeepSeek-R1's context window?

DeepSeek-R1 has a context window of 128K ctx.

How much does DeepSeek-R1 cost?

DeepSeek-R1 is priced at $0.55 per 1M input tokens and $2.19 per 1M output tokens.

DeepSeek-R1

Open weights

Reasoning DeepSeek Docs ↗

Model spec

Provider: DeepSeek
Context window: 128K ctx
Max output: -
Input price: $0.55 / 1M tokens
Output price: $2.19 / 1M tokens
License: MIT
Open weights: Yes

DeepSeek-R1 is a reasoning-focused AI model designed for tasks that demand extended, logical thought processes. Its reinforcement learning training and substantial context window differentiate it from general-purpose language models, positioning it for analytical workloads requiring deep textual comprehension. As an open-weights model under MIT license, it appeals to developers needing transparency or customization capabilities unavailable with proprietary alternatives. Organizations with reasoning-intensive text applications—particularly those requiring scrutiny of long documents or generation of methodical explanations—should evaluate this model, though they may want to consider output costs in high-volume scenarios.

Modality

Text

Use cases

Multi-step logical reasoning tasks requiring detailed explanationsAnalyzing long documents with intricate dependenciesGenerating structured, step-by-step solutions for complex problemsEducational applications requiring thorough, reasoned responses