Text Generation
Transformers
Safetensors
English
llama
causal-lm
text-generation-inference
4-bit precision
gptq

Bravo!

#3
by Ukro - opened

I was testing non english language(can't write in details) and i must say the result in comparison with other 10 vicuna models that i have tested even yours vicuna-13B 1.1, this stable-vicuna give VERY GOOD result!!!
Thank you again :D :>

You're welcome! Glad it's working well.

Ukro changed discussion status to closed

could you add more detail about your good results ?
maybe we should come up with a test to rate these models...

ive mentioned in another thread that my results didnt blew me away.
asking for random funny facts

This is a good idea! I would like to see a few questions that are hard for the models to answer and a way to gauge them.

I can say that the language is non English and was doing a text rewrite with additional changes.
I am following this guy, he gives good comparison from my POV on youtube:
StableVicuna is NOW the UNSTOPPABLE 13B LLM KING! Bye Vicuna!
https://youtu.be/QeBmeHg8s5Y

Sign up or log in to comment