AI Briefing#mlx
The latest AI and developer news about #mlx, with the original source and a short summary.
The latest AI and developer news about #mlx, with the original source and a short summary.
Open-source KQuant packs for Qwen models on Hugging Face achieve up to 3.7x lower KL divergence than public 4-bit baselines, with 6-15% faster inference on M5 Pro for 4B, 9B, and 27B models.
The open-source library supports exact k-NN, k-means, NN-Descent, and CAGRA algorithms, outperforming CPU-based hnswlib in both build time and search speed.
AnySpeechRecognizer, a Swift package that reimplements the Apple Speech API to enable the use of open models, and AnyDecisionModel for decision tasks have been released.
Software Mansion's open source app Private Mind now supports LFM3 models, achieving ~40 tok/s on iPhone 16 Pro via a custom ExecuTorch export.
The packages wrap Apple's new SpeechAnalyzer API with NVIDIA and Alibaba models, achieving 1.55% WER on LibriSpeech test-clean, and enable typed decision outputs via six backends including MLX and llama.cpp.
The free Python package ports models like Cloudflare's Clef and Perplexity's pplx-decider to MLX, achieving sub-20ms latency on small encoders on an M5 Pro and 96.4% accuracy on TREC after fine-tuning.
The framework now supports serving LLMs via OpenAI, Anthropic, and Ollama protocols simultaneously using a unified .flm file format.
Underdog has released 'Husky', a model-specific inference engine that is faster than MLX across all measured metrics, achieving up to 4.5x performance gains particularly in editing tasks.
Hugging Face has hired oMLX developer Jun Kim to strengthen support for the Apple Silicon-based MLX ecosystem and local AI infrastructure.
Open-source tool released that enables cloud AI agents to control local Piper for cost-effective code generation
The INT4 quantized version of the NVIDIA Cosmos3 64B model has been released for CUDA and MLX, enabling local execution.
An abliterated version of Qwen3.8-27B with censorship removed has been released in MLX format for Apple Silicon.
Released an MLX quantized version of Qwen3.8-27B with safety alignment removed for Apple Silicon.
Logue, a privacy-focused meeting notes app that runs 100% on-device using MLX on Apple Silicon, has been open-sourced.
A technology was unveiled that performs RL post-training using 14 Macs distributed across 4 countries worldwide.
An ExecuTorch integration has been released that lets React Native apps run Gemma 4 offline.
Apple has released MLX LM Server, which supports Continuous Batching and distributed inference.
Unsloth Studio now supports model training using MLX on Mac.
The SDK 'Cider,' which supports W8A8 activation quantization on Apple Silicon by addressing MLX's limitations, has been released.
An MLX port has been implemented that allows Cohere's 218B MoE model, Command A+, to run on Apple Silicon.
Native MTP support for Qwen3.6 MLX models has been added to the distributed inference tool exo.
A menu bar app has been released that lets you explore Hugging Face models on Mac and manage them with Ollama, MLX, and more.
It's an MLX-based local AI engine that supports inference up to 4.2x faster than Ollama on Apple Silicon.
Triggered by GitHub Copilot's billing shift, local LLM hardware is proposed as a solution for AI self-sufficiency.