Back to directory

MLXChat

An iPhone and iPad app that runs Alibaba's Qwen3.5 models locally on the device using Apple's MLX framework, with optional tool use, image input and web search.

MLXChat

MLXChat is a SwiftUI application for running Qwen3.5 language models directly on an iPhone or iPad through Apple's MLX framework, rather than calling a cloud API. It defaults to a 4-bit quantized 4-billion-parameter model, with smaller and larger MLX variants available depending on the device, and supports streaming chat, an optional reasoning mode, image input for compatible models, and downloading additional models from Hugging Face with a progress indicator and automatic retry.

Optional tools let the model fetch a URL directly or search the web through the Brave Search API for current information. A separate macOS command-line harness lets the same prompts and tools be iterated on outside the phone interface. It requires Apple Silicon and a recent Xcode to build, and targets recent iOS releases.

More in Inference

See all 770
  • OpenDecision

    deepanwadhwa.github.io

    An open-source Python package that answers typed questions about application state or documents using a local natural-language-inference model, returning structured choice, score, yes/no or evidence-relation values instead of free text.

  • Lifeboat

    iterate.ai

    A local inference server that puts an OpenAI- and Anthropic-compatible API in front of open-weight language models running on hardware you control, from a laptop to a GPU cluster.

  • Claude S40

    github.com

    An unofficial Claude chat client for 2007-era Nokia Series 40 phones, built as a Java ME app plus a small Go relay server, with web search, reading mode and saved chats.