← Back to Models Library
🤖

deepseek-r1

REASONING LEADER
📥 4.2M Downloads
🕒 1 month ago
#thinking #math #code #open-weights

DeepSeek-R1 achieves state-of-the-art reasoning performance on math, code, and logical tasks using pure reinforcement learning without supervised fine-tuning.

🏷️ Pricing Benchmark

Per 1M Tokens
Input
$0.55
Cached Input
$0.14
Output
$2.19

⚡ Quick Start Snippets

ollama run deepseek-r1

🚀 Applications

Ollama CLI

ollama run deepseek-r1

Local interactive terminal chat session with full chain-of-thought visible.

Open WebUI

docker run -d -p 3000:8080 ghcr.io/open-webui/open-webui:main

ChatGPT-like Web UI connected to local DeepSeek-R1 instance.