237 dependents
| Package | Description | Downloads/month |
|---|---|---|
| Run LLMs with MLX | 1.5M | |
| MLX-VLM is a package for inference and fine-tuning of Vision Language Models (VL... | 349K | |
| Examples in the MLX framework | 117K | |
| A text-to-speech (TTS), speech-to-text (STT) and speech-to-speech (STS) library ... | 78K | |
| MLX-Embeddings is the best package for running Vision and Language Embedding mod... | 51K | |
| MLX native implementations of state-of-the-art generative image models | 40K | |
| vMLX - Home of JANG_Q - Cont Batch, Prefix, Paged, KV Cache Quant, VL - Powers M... | 36K | |
| A high-performance API server that provides OpenAI-compatible endpoints for MLX ... | 22K | |
| The fastest local AI engine for Apple Silicon. 4.2x faster than Ollama, 0.08s ca... | 18K | |
| Fine-tune LLMs on your Mac with Apple Silicon. SFT, DPO, GRPO, Vision, TTS, STT,... | 14K | |
| An implementation of the Nvidia's Parakeet models for Apple Silicon using MLX. | 11K | |
| vLLM-like inference for Apple Silicon - GPU-accelerated Text, Image, Video & Aud... | 10K | |
| Core API and plugin system for LEANN | 7K | |
| Probably Hardly Ever Works — search-based superoptimizer for MLX/Metal | 7K | |
| InstructLab Core package. Use this to chat with a model and execute the Instruc... | 7K | |
| port of mlx-video but attempting to get audio ported as well finally | 6K | |
| Lumerico's Comprehensive Interface for Deep Learning | 6K | |
| Mixed-precision quantization optimizer for MLX models on Apple Silicon | 6K | |
| Timeseries signal processing implementations in ezmsg | 5K | |
| Implementation of F5-TTS in MLX | 5K | |
| Reusable mid-level building blocks for MLX — causal convolutions, multi-dim RoPE... | 5K | |
| Moshi is moshi, but running on macOS | 5K | |
| Qwen3-ASR speech recognition on Apple Silicon via MLX | 5K | |
| Merlin — a fast local LLM for agentic coding on Apple Silicon | 5K | |
| Implementation of 'Vocos: Closing the gap between time-domain and Fourier-based ... | 4K | |
| AlienSky optimized inference engine for Qwen on Apple Silicon | 4K | |
| Asynchronous Self-Healing KV Cache for Silicon-Native LLMs by GDI Nexus | 3K | |
| 3K | ||
| Lossless DFlash speculative decoding for MLX on Apple Silicon | 3K | |
| Extreme weight and KV cache compression for LLMs on Apple Silicon (MLX implement... | 3K | |
| Train LLMs on Apple silicon with MLX and the Hugging Face Hub | 3K | |
| Phi-3.5 for Mac: Locally-run Vision and Language Models for Apple Silicon | 3K | |
| Z-Vision Generator — cross-platform AI image and video generator | 2K | |
| On-device Image Generation for Apple Silicon | 2K | |
| Qwen3 CustomVoice 命令行工具 | 2K | |
| RETIX - The Optic Nerve for Autonomous Agents | 2K | |
| Unified MLX server & CLI (language and vision) with OpenAI-compatible endpoints | 2K | |
| Extraordinary speed, extraordinary quality — an MLX-based inference engine for A... | 2K | |
| Local speech synthesis for Apple Silicon — TTS, voice cloning, dialogue, and sou... | 2K | |
| Python utility for text embeddings in MLX. | 2K | |
| HuggingFace model management for MLX on Apple Silicon | 2K | |
| CanViT inference on Apple Silicon via MLX | 2K | |
| LLM fine-tuning & RL framework for MLX. | 2K | |
| YOLO26 implementation using Apple MLX framework | 2K | |
| Run AI models too large for your Mac's memory — expert caching, speculative exec... | 1K | |
| MLX + Metal implementation of mHC: Manifold-Constrained Hyper-Connections by Dee... | 1K | |
| Python tools for text to speech (TTS), speech to text (STT), and speech to speec... | 1K | |
| Metal-accelerated Vision Mamba for Apple Silicon (2D/3D/4D) with 3.9x training s... | 1K | |
| 3 AI models. 161B parameters. One Mac. 5.5GB. Full agentic pipeline on Apple Sil... | 1K | |
| Metal Flash Attention for MLX | 1K |
