danigoncalves

joined 10 months ago
[–] danigoncalves@alien.top 1 points 10 months ago

Best ones I tried were zephyr, dolphin, openorca, synthia and naberius.

[–] danigoncalves@alien.top 1 points 10 months ago

I was actually today comparing both (codellama 7B) and man codellama just gave crap, deepseek was vey accurate.

[–] danigoncalves@alien.top 1 points 10 months ago

All models are designed on a very similar way (fine tuned ones are even on a top level in this regard) if you are expecting Jarvis like model than I am sorry because I think Tony Stark didn't born yet 😅 Try having another perspective about the models and its limitations and you could actually start being comfortable with what open source give to our community.

[–] danigoncalves@alien.top 1 points 10 months ago

What would be the equivalent models based on open source and free for commercial use?

[–] danigoncalves@alien.top 1 points 10 months ago

Congrats for the launch! Its nice to witness the rise of these new open source tools.

 

Hello mates!

While working with local modals mostly on software development, I was wondering what would be the best model to work in my native language (Portuguese). Until now I never needed that kind of quality (language specific trained model), but I was thinking if I would need, what would be the model to have my hands on (and even further fine tune it). I know a llama based fine tuned model (Cabrita) but its a little bit restrictive in terms of usage, does anyone know any other llama or mistral model trained or fine tuned in portuguese?