News for hackers. Stuff that matters.

A comment on Gemma 3 270M: The compact model for hyper-efficient AI. Parent comment

Re:Gemma 3 270M: The compact model for hyper-efficient AI by wanderingmind on Friday August 15, 2025 @06:41AM (#44910707)
Why should we start fine tuning gemma when it is so bad. Why not instead focus the fine-tuning efforts on Qwen, when it starts off with much, much better outputs?
[ Parent | on HN ]
Re:Gemma 3 270M: The compact model for hyper-efficient AI by mdp2021 on Friday August 15, 2025 @08:14AM (#44911425)
Speed critical applications, I suppose. Have you compared the speeds?

(I did. I won't give you number (which I cannot remember precisely), but Gemma was much faster. So, it will depend on the application.)

[ Parent | on HN ]