Comment by PlatoIsADisease
6 hours ago
If it worked IRL for anything useful, I'd be more interested in the technical differences. But it was a mere toy for a few tests at my fortune 20 company.
Language is full of issues of particulars vs universals, and you could debate if its just an integrated GPU with different marketing.
Whatever the case, we couldn't use it in production, and NVIDIAs stock price reflects the reality on the ground.
Well, I've been using a fine-tuned variant of Gemma 3 model since it came out, and some embedding models, on a laptop. It's not "useless" by any means, in fact it still beats the latest Claude for my use-case in Ukrainian. Not to mention that if you travel by train a lot, you will find it quite useful. I own a Mac studio M2 Max (96 GB) variant at home, and I'm routinely using the larger models for the kind of stuff I don't wish to share with model providers.
My 2 cents