Model Releases

Gemma 4 12B released: 150M+ downloads milestone, runs locally on 16GB VRAM

Original title:Celebrating the milestone of a massive 150+ million downloads of Gemma 4 with the release of the new...

AI summary

Demis Hassabis announced that the Gemma 4 family has surpassed 150 million downloads and officially released the new Gemma 4 12B model. The model is a unified, encoder-free multimodal model that combines edge efficiency with advanced reasoning. Despite having only 12B parameters, it delivers strong performance and is small enough to run locally on a laptop with just 16GB of VRAM. It is released under the Apache 2.0 open-source license, making it easy for developers to build freely.

Key Takeaways

  • Demis Hassabis announced that the Gemma 4 family has surpassed 150 million downloads and…
  • The model is a unified,encoder-free multimodal model that combines edge efficiency with advanced…
  • Despite having only 12B parameters,it delivers strong performance and is small enough to run…
  • It is released under the Apache 2.
  • 0 open-source license,making it easy for developers to build freely.

Quick Facts

Published
Updated
Source
X:Demis Hassabis (@demishassabis)
Category
Model Releases
AI score
74
Read the original Read the full article at the original source
Why it matters

Gemma 4 12B 用 Apache 2.0 许可把多模态模型压进笔记本,16GB 显存就能跑,端侧智能的性价比又一次被 Google 拉高,做本地推理的可以马上试试。

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: Gemma 4 12B released: 150M+ downloads milestone, runs locally on 16GB VRAM. June 4, 2026. https://new.nat6.net/en/news/cmpyerfzl04g6slaxbcd0ly8n/

View all →