minimal-llm-ui-demo.mp4
This minimalistic UI is designed to act as a simple interface for Ollama models, allowing you to chat with your models, save conversations and toggle between different ones easily. The tool is built using React, Next.js, and Tailwind CSS, with LangchainJs and Ollama providing the magic behind the scenes.
- Chat with Local Language Models (LLMs): Interact with your LLMs in real-time through our user-friendly interface.
- Model Toggling: Switch between different LLMs easily (even mid conversation), allowing you to experiment and explore different models for various tasks.
- Memory-based Context Storage: Keep track of context in memory, ensuring smooth interactions even when switching between models.
- Storing conversation history.
- Built using React, Next.js, and Tailwind CSS for a clean and modern design.
- Utilizes LangchainJs and Ollama for seamless integration with Local Language Models (LLMs).
- Stores context in memory for efficient model switching.
- Download and run Ollama on your machine with
ollama serve
orollama run <model-name>
(it will run at: http://localhost:11434/) - Open a new terminal and navigate to the root of this project.
- Install the dependencies
npm install
in your terminal. - Start the tool by running
npm run dev
(it should be available in your web browser athttp://localhost:3000
)
- Add edit message icon for user messages
- Summarise conversations
- Incorporate visualisations
- Convert to desktop app so that it can be more powerful
- Allow for prompt savings + shortcuts
If you encounter any issues, feel free to reach out!
This project is licensed under the MIT License. See LICENSE
file for details.