DeepChat is a local chatbot project integrated with a modern web interface, developed with React.js. The system utilizes the DeepSeek-R1 artificial intelligence model, processed via Ollama.
- Clone the Repository
Run the following command in the terminal to clone the project:
git clone https://github.com/Dnowdd/DeepChat.git
cd DeepChat- Install Dependencies
Make sure you have Node.js installed on your machine. Then, install the project dependencies by running:
npm install- Configuring .env
Before running the project, check if the .env file exists at the root of the project. If it doesn't, you can create and configure it by providing information such as
AI_ENDPOINTandAI_MODELso that the system can communicate with DeepSeek.
VITE_AI_ENDPOINT=http://localhost:11434/api/generate # AI API endpoint, the ollama default is http://localhost:11434/api/generate
VITE_AI_MODEL=deepseek-r1:8b # AI model name. The project was built using deepseek-r1:8b, but you can change it to any other model- Run the Project
npm run dev- Run Deepseek
To use DeepSeek-R1, first install Ollama. After installation, open a new terminal and execute the following command:
ollama run deepseek-r1:8bThe 8b parameter refers to a model with 8 billion parameters. The higher the number of parameters, the greater the computational resource requirements, but also the more advanced the artificial intelligence will be. For more information on available models, refer to the official DeepSeek documentation on the Ollama website.
- Switch between website languages:
en-usandpt-br - Cancel AI text generation at any time
- Fully responsive interface
- Support for multiple chat sessions
- Custom canvas for code snippets
- Markdown support for text formatting
- Enhanced input prompt for improved user experience
We encourage contributions from the open-source community. Feel free to submit issues, suggestions, or pull requests to help improve DeepChat.
DeepChat is distributed under the MIT License, allowing for open collaboration and modification.
For any inquiries, suggestions, or feedback, don’t hesitate to reach out:
- Github: Dnowdd
- LinkedIn: Dnowdd
- Email: david.queiroz@dnowdd.com


