SmolChat

Shubham Panchal

About this app

SmolChat allows you to download and run popular LLMs on your Android device, locally, without needing an internet connection. Customize the model used for each chat, tune settings like temperature and min-p, and pin your favourite chats on the home-screen with shortcuts.

Ratings and reviews

Looks promising but didn't work for me. Trying to use qwen, messages after the first fail with: Unknown built in filter tojson in the template, and I've no idea what the fix is. Attempting to use Gemma throws an error that conversational files must alternate. Phi 4 gave 2 coherent messages then garbage after which the app crashed and won't reopen. Clearing data and cache fixed that subsequent responses were unrelated to the prompt.

overall I think a good app that allows you to play with any gguf file from hugging face with minimal fuss. I was a bit disappointed to see that it doesn't handle reasoning models quite as well as I had hoped, in that it displays reasoning/thinking tokens as if they were normal tokens. So the thinking is indistinguishable from the response which is unhelpful.

Um, by default, SmolChat has NO File permissions, so it can not load the LLM file from the Download directory. Help? Permission manager has it strictly locked down and locked out?

What’s new

- Sync with upstream llama.cpp brings in support for Gemma 4 models

- 'copy' and 'share' buttons below a message, now, do not copy the thinking contents of the model

- Fix a bug where the app crashes when trying to show a list of available models from HuggingFace in the absence of an internet connection