The short version
Google’s open-weight Gemma 4 model family for local inference.
From the HackerLinks archive
Gemma 4
Google’s open-weight Gemma 4 model family for local inference.
At a glance:
First seen:2026-05-10
Last seen:2026-05-10
Times seen:1
Website:deepmind.google
Why it caught our attention
Commenters said it was fast enough and useful for private local search tasks.
Where it surfaced on Hacker News
2026-05-10
Editorial paraphrase
One commenter said they were using gemma4 locally on an MBP M2 and it felt really good; another called Gemma4 crazy well trained for its size.
Original threadLocal AI needs to be the norm