Delta LM is text input/text output app that allows you to chat with language models with all the processing happening on your device itself. All your chat is saved on your device. The only time internet is required in the app is when user wants to download language models from the provided options. The context of the Language models is limited to the current chat session only and the LMs will not be able to recall any previous chats form previous sessions. The app uses GGUF-format models with Q4KM quantization, which will be saved to your device Downloads folder. This app is a modified version of an open-source project by Andriy Druk.


Currently supported models:

• IBM Granite 3.0 2B and 8B
• Meta Llama 3.2 1B and 3B
• Qwen 2.5 0.5B, 1.5B, 3B and 7B
• Qwen 2.5 Coder 1.5B, 3B and 7B
• Google Gemma2 2B and 9B
• Microsoft Phi3.5
• Llama 3.1 8B