MNN Chat

jindesong

About this app

This is a multimodal language model (LLM) Android app

Multimodal Support: Enables functionality across diverse tasks, including text-to-text, image-to-text, audio-to-text, and text-to-image generation (via diffusion models).

CPU Inference Optimization: MNN-LLM demonstrates exceptional performance in CPU benchmarking in Android.

Broad Model Compatibility: Supports multiple leading model providers, such as Qwen, Gemma, Llama (including TinyLlama and MobileLLM), Baichuan, Yi, DeepSeek, InternLM, Phi, ReaderLM, and Smolm.

Privacy First: Runs entirely on-device, ensuring complete data privacy with no information uploaded to external servers.

Ratings and reviews

Make it structured more like a conversation-based app in which the usual layout on most AI apps can be seen, with conversations, a new chat button on the sidebar and a library with your models. Also, allow the options that most cloud AI apps (and some local ones) have already adopted, including an edit button, a regeneration button and other things.

nice app but I am finding some crashes on occasion. what would make this amazing is to be able to upload a model from a device and the ability to export and import chats in case you switch device.

Very fast. BUT buggy mess, specially with qwen models.I tried feeding the model with a prompt file with image path, but the model produces gibberish or repetitive letters. Only when I changed "memory": from low to normal the model finally worked. The downside is it became super slow. Please help. My device is poco x3 gt 8gb lpddr4, mali g77 mc9, dimensity 1100.

What’s new

add gemma4 and lfm2 support