AISimon Willison4h ago
Gemma 4 audio with MLX
TL;DRGemma 4 can now transcribe audio on Mac using MLX framework.
Why it matters: Enables efficient on-device audio processing without cloud dependencies or expensive hardware.
Thanks to a tip from Rahim Nathwani, here's a uv run recipe for transcribing an audio file on macOS using the 10.28 GB Gemma 4 E2B model with MLX and mlx-vlm: uv run --python 3.13 --with mlx_vlm --with torchvision --with gradio \ mlx_vlm.generate \ --model google/gemma-4-e2b-it…
Read full articleSource: Simon Willison · Opens in new tab