Ollama GUI is a web interface for ollama.ai, a tool that enables running Large Language Models (LLMs) on your local machine.
- Download and install ollama CLI.
ollama pull <model-name>
ollama serve
- Clone the repository and start the development server.
git clone https://github.com/HelgeSverre/ollama-gui.git
cd ollama-gui
yarn install
yarn dev
Or use the hosted web version, by running ollama with the following origin command (docs)
OLLAMA_ORIGINS=https://ollama-gui.vercel.app ollama serve
For convenience and copy-pastability
, here is a table of interesting models you might want to try out.
For a complete list of models Ollama supports, go to ollama.ai/library.
Model | Parameters | Size | Download |
---|---|---|---|
Mistral | 7B | 4.1GB | ollama pull mistral |
Mistral (instruct) | 7B | 4.1GB | ollama pull mistral:7b-instruct |
Llama 2 | 7B | 3.8GB | ollama pull llama2 |
Code Llama | 7B | 3.8GB | ollama pull codellama |
Llama 2 Uncensored | 7B | 3.8GB | ollama pull llama2-uncensored |
Orca Mini | 3B | 1.9GB | ollama pull orca-mini |
Vicuna | 7B | 3.8GB | ollama pull falcon |
Vicuna | 7B | 3.8GB | ollama pull vicuna |
Vicuna (16K context) | 7B | 3.8GB | ollama pull vicuna:7b-16k |
Vicuna (16K context) | 13B | 7.4GB | ollama pull vicuna:13b-16k |
nexusraven | 13B | 7.4gB | ollama pull nexusraven |
starcoder | 7B | 4.3GB | ollama pull starcoder:7b |
wizardlm-uncensored | 13B | 7.4GB | ollama pull wizardlm-uncensored |
- Properly format newlines in the chat message (PHP-land has
nl2br
basically want the same thing) - Store chat history using IndexedDB locally
- Cleanup the code, I made a mess of it for the sake of speed and getting something out the door.
- Add markdown parsing lib
- Allow browsing and installation of available models (library)
- Ensure mobile responsiveness (non-prioritized use-case atm.)
- Add file uploads with OCR and stuff.
- Ollama.ai - CLI tool for models.
- LangUI
- Vue.js
- Vite
- Tailwind CSS
- VueUse
- @tabler/icons-vue
Licensed under the MIT License. See the LICENSE.md file for details.