/AI9h ago

MLX-VLM Optimizes Gemma 4 12B For Local Apple Silicon Inference

--0--
Quote posts
Dmitry Lyalin
Reposts
Olivier Bachem#1188
Original postOlivier Bachem#1188
Dmitry Lyalin@LyalinDotCom

First, amazing to see MLX optimization launch! second, Prince does a great job highlighting the top model features. You get SO much with this model.

Prince Canuma@Prince_Canuma

🚀 Gemma 4 12B is here!

We partnered with @GoogleDeepMind to bring and optimize their new dense and unifed multimodal model for Apple Silicon.

◈ 12B dense · 256K context ◈ Thinking mode (built-in reasoning) ◈ Vision: dynamic res, OCR, UI + charts ◈ Native audio: ASR + speech translation ◈ Function calling for agents ◈ Text + image + audio, interleaved

Runs local. Get started now ⚡

> uv pip install -U mlx-vlm

https://github.com/Blaizzy/mlx-vlm

11:44 AM · Jun 3, 2026 · 3.1K Views
Sentiment
Sentiment building, check back later.
Cluster Engagement
-
Views
-
Comments
-
Reposts
-
Bookmarks
Expand data
Posts from X
Most Activity
Most ActivityTimeline
RETWEETS47
Prince Canuma@Prince_Canuma

🚀 Gemma 4 12B is here!

We partnered with @GoogleDeepMind to bring and optimize their new dense and unifed multimodal model for Apple Silicon.

◈ 12B dense · 256K context ◈ Thinking mode (built-in reasoning) ◈ Vision: dynamic res, OCR, UI + charts ◈ Native audio: ASR + speech translation ◈ Function calling for agents ◈ Text + image + audio, interleaved

Runs local. Get started now ⚡

> uv pip install -U mlx-vlm

https://github.com/Blaizzy/mlx-vlm

Google Gemma@googlegemma

Meet Gemma 4 12B!

A unified, encoder-free multimodal model designed to bring high-performance intelligence directly to your laptop, and released under an Apache 2.0 license.

Bridging the gap between edge efficiency and advanced reasoning. Here is what’s new with Gemma 4 12B: 👇

10hViews 83.7KLikes 929Bookmarks 609