awesome-mlx

KO | EN

gitlite — search

#javascript #hacktoberfest #react #typescript #golang #android #rust #docker #macos #nodejs #windows #awesome #pytorch #mcp

(★ 241)

No description available.

File Explorer

Download Latest Version (.zip)

README.md

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Awesome MLX

An awesome list dedicated to the [MLX library](https://github.com/ml-explore/mlx) from Apple

## Articles

### 2024
- [On-device ML research with MLX and Swift](https://www.swift.org/blog/mlx-swift/)
- [MLX: On-device machine learning on Apple Silicon](https://antran.app/2024/mlx_getting_started/)
- [A Deep Dive into Apple’s Machine Learning Framework (MLX) — Step by Step Introduction](https://medium.com/@dynotes/a-deep-dive-into-apples-machine-learning-framework-mlx-step-by-step-introduction-d00681e56de2)
- [Deploying LLMs locally with Apple’s MLX framework](https://towardsdatascience.com/deploying-llms-locally-with-apples-mlx-framework-2b3862049a93)
- [MLX Tutorial: Apple's Bold Move in Machine Learning Arena!](https://www.youtube.com/watch?v=DIpv7CQloSk) 
- [Deep Dive into AI with MLX and PyTorch](https://github.com/neobundy/Deep-Dive-Into-AI-With-MLX-PyTorch): "Deep Dive into AI with MLX and PyTorch" is an educational initiative designed to help anyone interested in AI, specifically in machine learning and deep learning, using Apple's MLX and Meta's PyTorch frameworks.
- [Notes on the Apple MLX machine learning framework](https://github.com/uogbuji/mlx-notes): Shared personal notes created while working with the Apple MLX machine learning framework

## Models

- [Models on HuggingFace](https://huggingface.co/mlx-community)
- [Using MLX at Hugging Face](https://huggingface.co/docs/hub/en/mlx)
- [llama-3-8B-Instruct-function-calling](https://huggingface.co/mzbac/llama-3-8B-Instruct-function-calling): This model is fine-tuned based on Meta-Llama/Meta-Llama-3-8B instructions via mlx-lm.

## Libraries and Tools

- [atlas](https://github.com/Matth21/atlas): Measured-cost LLM quantizer — profiles per-block KL sensitivity, then solves (bit-width, group-size) allocation exactly for any RAM budget (`atlas <model> --budget-gb 6`). Ships pre-computed cost tables; runs entirely on-device.
- [mlxserver](https://www.mlxserver.com/): This Python library is the easist way to begin building on top of Apple's machine learning library, MLX.
- [mlx-llm](https://github.com/riccardomusmeci/mlx-llm): Large Language Models (LLMs) applications and tools running on Apple Silicon in real-time with Apple MLX.
- [lightning-whisper-mlx](https://github.com/mustafaaljadery/lightning-whisper-mlx): An incredibly fast implementation of Whisper optimized for Apple Silicon.
- [mlx-benchmark](https://github.com/TristanBilot/mlx-benchmark): Benchmark of Apple MLX operations on all Apple Silicon chips (GPU, CPU) + MPS and CUDA.
- [MLX-VLM](https://github.com/Blaizzy/mlx-vlm): MLX-VLM is a package for running Vision LLMs locally on your Mac using MLX.
- [mlx-llm-server](https://github.com/mzbac/mlx-llm-server): For inferring and serving local LLMs using the MLX framework
- [mlx-transformers](https://github.com/ToluClassics/mlx-transformers): MLX Transformers is a library that provides model implementation in MLX. It uses a similar model interface as HuggingFace Transformers and provides a way to load and run models in Apple Silicon devices.
- [mlxcli](https://github.com/mustafaaljadery/mlxcli): Run large models from the terminal using Apple MLX.
- [MFlux](https://github.com/mflux-community/mflux):  MLX native implementations of state-of-the-art generative image & video models
- [SiLLM](https://github.com/armbues/SiLLM): SiLLM simplifies the process of training and running Large Language Models (LLMs) on Apple Silicon by leveraging the MLX framework.
- [MLX_CLIP 📚🤖](https://github.com/harperreed/mlx_clip): A simple package to use CLIP on apple silicon using the MLX libraries from Apple
- [outlinesmlx](https://github.com/sacha-ichbiah/outlines-mlx): A fast minimalistic implementation of guided generation on Apple Silicon using Outlines and MLX
- [mlx-embedding-models](https://github.com/taylorai/mlx_embedding_models): Run text embeddings on your Apple Silicon GPU. Supports any BERT- or RoBERTa-based embedding model, with a curated registry of high-performing models that just work off the shelf.
- [RLX: Reinforcement Learning with MLX](https://github.com/noahfarr/rlx): https://github.com/noahfarr/rlx
- [🍏 MLX - ResNet 🍏](https://github.com/Aavache/mlx-resnet): ResNet implementation with the MLX, Apple's DL framework.
- [mlx_stft](https://github.com/nuniz/mlx_stft): Short-Time Fourier Transform (STFT) for MLX (Apple)
- [PyOllaMx (Ollama + MlX)](https://github.com/kspviswa/pyOllaMx): Your gateway to both Ollama & Apple MlX models
- [mlx-bench](https://github.com/argmaxinc/mlx-bench): Benchmark MLX Performance across commits
- [mlx-image](https://github.com/riccardomusmeci/mlx-image): mlx image models for Apple Silicon machines
- [mlx-local-server](https://github.com/Doriandarko/mlx-local-server): A tiny server to run local inference on MLX model in the style of OpenAI
- [AutoMLX](https://github.com/wsvn53/AutoMLX): An easy-to-use LLMs inference tool for quickly loading models accelerated by the Apple MLX framework on Mac devices, and providing a simple and compatible API interface for integration with other LLMs tools.
- [mlx-omni-server](https://github.com/madroidmaq/mlx-omni-server): MLX Omni Server is a local inference server powered by Apple's MLX framework, specifically designed for Apple Silicon (M-series) chips. It implements OpenAI-compatible API endpoints, enabling seamless integration with existing OpenAI SDK clients while leveraging the power of local ML inference.
- [Rapid-MLX](https://github.com/raullenchai/Rapid-MLX): OpenAI-compatible LLM inference server for Apple Silicon. 2-4x faster than Ollama with full tool calling, reasoning separation, and prompt caching.
- [PostTrainLLM](https://github.com/PostTrainLLM/posttrainllm): Mac-local factory for post-training, evaluating, and packaging specialist LLMs with MLX and evidence-backed evaluation gates.
- [mlx-local-server](https://github.com/Ar9av/mlx-lm-server): OpenAI-compatible inference servers for Apple Silicon built in Rust — covers LLM (chat, embeddings, vision, LoRA, fine-tuning), image generation (FLUX.2-klein, 9s/image), and audio (TTS, STT, source separation). 8MB idle RAM, 16ms cold start.
- [mlx-serve](https://github.com/ddalcu/mlx-serve): Native Zig inference server for Apple Silicon. OpenAI- and Anthropic-compatible APIs (works with Claude Code), runs both MLX and GGUF models, with speculative decoding and KV-cache quantization. Ships MLX Core, a macOS menu-bar app with chat, agent mode, and tool calling.

- [docker_mlx_cpp](https://github.com/RobotFlow-Labs/docker_mlx_cpp) - The NVIDIA Container Toolkit for Mac. Give any Docker container full Metal GPU access. 107+ GPU operations via MLX.
- [mlx-teacache](https://github.com/IonDen/mlx-teacache): TeaCache step-skipping wrapper for mflux FLUX models on Apple Silicon, with per-variant benchmark notes.
- [mlx-taef](https://github.com/IonDen/mlx-taef): TAESD/TAEF tiny autoencoders in MLX for fast diffusion latent previews and low-memory decode on Apple Silicon.
- [mlx-sparse](https://github.com/waleed-sh/mlx-sparse): Sparse array containers, primitives and linear algebra for MLX.
- [MacProvider](https://github.com/Augustas11/macprovider): Coordinator + gateway that turns any Apple Silicon Mac into a remote-addressable mlx-lm endpoint. OpenAI-compatible API, outbound-WebSocket provider daemon (no port-forwarding), and signed receipts binding (prompt, output, provider) for verifiable inference.

## Demo

- [MFlux](https://github.com/mflux-community/mflux):  MLX native implementations of state-of-the-art generative image & video models
- [flux-generator](https://github.com/voipnuggets/flux-generator): Flux Generatoe is a high-performance macOS web application that enabled local image generation using Blackforest Flux models (Schnell/Dev) and Stable Diffusion (sdxl-turbo/stable-diffusion-2-1) and musci generation using facebook/musicgen. It also provides an API for image generation compatable with third party applications like open-webui.
- [mlx-chat-app](https://github.com/mlx-chat/mlx-chat-app): Chat with MLX is a high-performance macOS application that connects your local documents to a personalized large language model (LLM).
- [photo-similarity-search](https://github.com/harperreed/photo-similarity-search): Super simple MLX (apple silicon) CLIP based photo similarity web app
- [Video Summarization](https://github.com/Peter-obi/Video_summarization_mlx): Transcribe and summarize videos using whisper and llms on apple mlx framework
- [MLX Assistant](https://github.com/Doriandarko/mlx-assistant): A simple script to enhance text editing across your Mac, leveraging the power of MLX. Designed for seamless integration, it offers real-time text correction everywhere you type.
- [MLX RAG With GGUF Model Weights](https://github.com/Jaykef/mlx-rag-gguf): Minimal, clean code implementation of RAG with mlx using gguf model weights
- [Enhancing Video Analysis with MLX: A Guide](https://github.com/abhishek-ch/mlx-video-qa): Explore the capabilities of the MLX library and leverage the genAI stack on MacOS to interact with any video.
- [MLX Demos](https://github.com/RAHB-REALTORS-Association/mlx-demos): Explore machine learning techniques with Gradio interfaces for Stable Diffusion image generation and LoRA text generation with the Apple MLX framework.
- [MLX Chat](https://github.com/da-z/mlx-ui): A simple UI / Web / Frontend for MLX mlx-lm using Streamlit.
- [SiLLM-examples](https://github.com/armbues/SiLLM-examples): Examples for using SiLLM
- [PyOMlx](https://github.com/kspviswa/PyOMlx): A wannabe Ollama equivalent for Apple MlX models
- [MLX Community Projects](https://github.com/ml-explore/mlx/discussions/654): Let's collect some cool MLX integrations and community lead projects here for visibility!
- [nanoGPT_mlx](https://github.com/vithursant/nanoGPT_mlx): Port of Andrej Karpathy's nanoGPT to Apple MLX framework.
- [Autogram](https://github.com/ivanfioravanti/autogram): Grammar checker with a keyboard shortcut for Ollama and Apple MLX with Automator on macOS.
- [MLX RAG](https://github.com/AbeEstrada/mlx-rag): Retrieval Augmented Generation (RAG) example using Langchain (loader), Chroma (vector database), Hugging Face sentence-transformers (embeddings) and MLX (LLM).

* [Sumika](https://github.com/ngutech21/sumika-chat): Local-first macOS agent for private AI workflows with chat, workspace context, and tool execution, running Gemma models with MLX/MLX Swift.

## MLX Swift
- [MLX Swift](https://github.com/ml-explore/mlx-swift): Swift API for MLX
- [MLX Swift Examples](https://github.com/ml-explore/mlx-swift-examples): Examples using MLX Swift
- [PrivateFoundationModels](https://github.com/john-rocky/PrivateFoundationModels): Drop-in API-compatible mirror of Apple's FoundationModels framework on iOS 18+. Routes `LanguageModelSession.respond(...)` to MLX-Swift (`mlx-community/*` LLMs + VLMs) via `PrivateFoundationModelsMLX`, or to CoreML / Apple's native FM on iOS 26+.
- [mlx-swift-chat](https://github.com/PreternaturalAI/mlx-swift-chat): A multi-platform SwiftUI frontend for running local LLMs with Apple's MLX framework.
- [ChatMLX](https://github.com/maiqingqiang/ChatMLX): ChatMLX is a large model real-time conversation app implemented using MLX
- [Ph3iOSOnDeviceChatApp](https://inkysquid4.gumroad.com/l/lghejp): Source code to run Microsoft's Phi 3 Min 4K model completely on device
- [fullmoon-ios](https://github.com/mainframecomputer/fullmoon-ios): fullmoon is an iOS app to chat with local large language models that’s optimized for Apple silicon and works on iPhone, iPad, and Mac. your chat history is saved locally, and you can customize the appearance of the app.
- [asiai](https://github.com/druide67/asiai): Multi-engine LLM benchmark & monitoring CLI for Apple Silicon. Compare MLX engines (LM Studio, mlx-lm, oMLX) against llama.cpp (Ollama) — tok/s, TTFT, power efficiency. Web dashboard and MCP server included.
- [Chaty](https://github.com/Fangyuan025/Chaty): Local, private desktop AI app (Windows/macOS) with a native MLX backend — a process-isolated Swift sidecar built on mlx-swift-lm — alongside GGUF/llama.cpp. Chat, RAG, vision, voice, and an agentic coding mode, fully offline. (MIT)
- [Logue](https://github.com/bitwize-ai/Logue): Privacy-first AI meeting-notes and writing assistant for macOS. Runs transcription, speaker diarization, and LLM inference on-device via MLX-Swift on Apple Silicon — nothing leaves your Mac. (MIT)

awesome-mlx

File Explorer

Today's picks

Was this content helpful?

Recent Feedback

awesome-mlx

File Explorer

Today's picks

Was this content helpful?

Recent Feedback

Download README