---
title: "DeepSeek R1 — LLM profile"
provider: "DeepSeek"
version: "r1"
model_type: "llm"
context_window: "131072"
parameters: "671B"
release_date: "2025-01-20"
url: "https://lforla.org/models/deepseek-r1"
markdown_twin: "https://lforla.org/markdown/models/deepseek-r1.md"
---
# DeepSeek R1

## Overview

DeepSeek-R1 is the first open reasoning model competitive with frontier closed models at a fraction of the cost.

## Capabilities
- Chain-of-thought reasoning
- Open weights
- Strong math and code

## References
- [DeepSeek-R1: Incentivizing Reasoning in LLMs](https://arxiv.org/abs/2501.12948)