Here's a detailed for setting up and running DeepSeek R1 locally with all the mentioned resources and instructions:
DeepSeek R1 is a powerful open-source AI model capable of performing reasoning, coding, and math tasks, comparable to OpenAI o1 and Claude 3.5 Sonnet. This guide will help you set up DeepSeek R1 locally on your machine for free and with complete privacy.
Ollama is a tool for running AI models locally on your machine.
- Download for Mac, Windows, or Linux
- Refer to the image below for downloading Ollama:
.png)
.png)
- Follow the installation instructions based on your OS.
- Example for installation:
.png)
.png)
DeepSeek R1 offers multiple sizes based on your hardware capability:
- 1.5B (Smallest):
ollama run deepseek-r1:1.5b - 8B:
ollama run deepseek-r1:8b - 14B:
ollama run deepseek-r1:14b - 32B:
ollama run deepseek-r1:32b - 70B (Largest):
ollama run deepseek-r1:70b
Start with a smaller model to test performance.
Open your terminal and run the following command:
ollama run deepseek-r1:8bRefer to the image for downloading progress:
It will download the model. Wait until it reaches 100%:
.png)
.png)
Chatbox is a clean, privacy-focused desktop interface for interacting with AI models.
- Chatbox Official Site
- Example for downloading on Windows:
.png)
.png)
- Open Chatbox and navigate to Settings.
- Set the Model Provider to Ollama.
- Configure the API Host to
http://127.0.0.1:11434.
Refer to the image below for configuration details:
.png)
- After completing the setup, you can start interacting with the model via Chatbox.
- Example of asking a question and receiving an answer:
.png)
.png)
- Ensure you have sufficient GPU resources for larger models (32B and 70B).
- For any issues, check the logs in Ollama or the Chatbox troubleshooting guide.
Enjoy using DeepSeek R1 for your AI tasks! Let us know if you have any feedback or questions. 🚀