Got a Mac with an M-chip? You can now train Gemma3 270m locally as a multilingual embedding or reranker model using our mlx-retrieval project. It lets you train Gemma3 270m locally at 4000 tokens/s on M3 Ultra - that's actually usable speed. We've implemented some standard practices for training an effective decoder-only embedding or reranker model with MLX: full/partial LoRA, InfoNCE, gradient accumulation, and streaming data loader. Plus MTEB integration for train-evaluation loops.

mlx-vlm v0.3.3 is here gr New models: - @LiquidAI_ LFM2-VL - @Zai_org GLM-4.5V - @cohere Command-A-Vision Changes: - New kernel for grid_sample - Fix bicubic interpolate kernel compatibility with macOS < 15 - Fix config inheritance Thank you very much to all the amazing friends at @lmstudio: @ostensiblyneil, @mattjcly and christian-lms. Get started: > pip install -U mlx-vlm Please leave us a star:

mlx-audio v0.2.4 is here 🔥🚀 New models: - IndexTTS - Voxtral Changes: # Python - Multi-model support for Visualizer - Fix deprecated save in MLX-LM - Fix mimi codec - Update CSM loading for TFMS style # Swift - Use custom URL for Kokoro - Troubleshooting guide - Add Misaki G2P tokenizer for Kokoro Thank you very much to all the amazing contributors: @lllucas, @ivanfioravanti, @adrgrondin, Senstella, Sam and Kinkadius! Get started: > pip install -U mlx-audio Please leave us a star ⭐:

Gemma 3 270M (8-bit) is also fast on iPhone 16 Pro 🏎️ ~140 tk/s for the A18 Pro chip with MLX, not that far from the M3 chip While Gemma 3 270M is not intended for chat, it's perfect to be used with Apple Shortcuts on tasks like summarization for example

Cool new MLX project: mlx-knife: an ollama like cli tool for MLX models on huggingface (pull, rm, list, show, serve, ...)

High altitude training makes LLM stronger

Who does not try training a classifier with MLX and Gemma 3 270m in flight?

Gemma 3 270M (8-bit) at ~200 tk/s on iPad Air M3 with MLX

In 44 lines of code, we have an actually useful agent that runs entirely locally, powered by Qwen3 30B A3B Instruct

Nvidia Parakeet v3 is out! Enjoy Day 0 support with Argmax SDK - What changed from v2? - How do I use it? - Should I upgrade to this model right away? Answers in comments