DeepSeek-R1 is an open-source reasoning model that you can run locally with Ollama. Here’s how to set it up.
Step 1: Install Ollama
Download from ollama.com or run curl -fsSL https://ollama.com/install.sh | sh on Linux/macOS.
Step 2: Pull DeepSeek-R1
Choose the size that fits your hardware:
ollama pull deepseek-r1:1.5b(lightweight)ollama pull deepseek-r1:7b(balanced)ollama pull deepseek-r1:14b(powerful)
Step 3: Run the Model
Start chatting with ollama run deepseek-r1:7b. For an API, use ollama serve and send requests to http://localhost:11434/api/generate.
Frequently Asked Questions
The 1.5B model runs on 8GB RAM; 7B needs 16GB, and 14B works best with 32GB or a GPU with 8GB+ VRAM.
Yes, Open WebUI automatically detects Ollama models and works great with DeepSeek-R1.
Yes, it’s released under the MIT license, so you can use it commercially.