open source is a hell of a drug. i dropped an ios port of gemma 3n yesterday, just a few days after google released it (with android-only support), and already someone else jumped in, added image understanding, polished the UX, and is prepping a PR. insane.

this is what it’s all about — no NDAs, no permission slips, just vibes and shipping.

gemma 3n is a frontier edge model, and rn we’re watching a legit community ecosystem form around it in real time. that’s magic.

huge thanks to @TheMagicIsInTheHole on reddit — here’s the screenshot they shared 👇

check out my reddit post for more discussion: 

https://reddit.com/r/LocalLLaMA/comments/1kvjwiz/implemented_a_quick_and_dirty_ios_app_for_the_new/

don't sleep on this subreddit, massive alpha on here!

i dabble with ai using open identity data from crypto https://alexpaden.tech

prev startups & racecars

oh interesting I guess this must be a new form of moe? I’m out of the loop but outside of this I think another huge exciting step is gluing together high speciality/niche models (I want to train on social data somehow, just not sure how yet)

Thanks for the comment + great work

it's even more insane than that.

"Gemma 3n models use selective parameter activation technology to reduce resource requirements. This technique allows the models to operate at an effective size of 2B and 4B parameters, which is lower than the total number of parameters they contain."

I’m still kind of baffled this can load into ram on a phone isn’t it like 4B param? I’ll have to read up but wow that image support is insane

nope, I don't think google has made that available yet

does this support audio too? i haven’t figured out how to do audio input and output in gemma 3n

I've been wildly impressed with Gemma3 in general.