semperai / amica

Amica is an open source interface for interactive communication with 3D characters with voice synthesis and speech recognition.
https://heyamica.com
MIT License
625 stars 96 forks source link

Model on online demo #3

Closed Aspie96 closed 8 months ago

Aspie96 commented 8 months ago

Hi.

It's not clear to me what model is being used in the online demo, but it appears to be LLaMA 2, correct me if I am wrong.

If this is so, may I suggest considering switching to something like mistral-openorca, zephyr, dolphin2.2.1-mistral, openhermes2.5-mistral, openchat or neural-chat?

Unlike LLaMA (2), which has a proprietary license by Meta, these models are all under an open source license and are based on Mistral, which is under the Apache 2.0 license and some of them may even outperform LLaMA 2.

kasumi-1 commented 8 months ago

Hi! The demo is using OpenChat 3.5 by Alignment Lab, which is Mistral based.

Aspie96 commented 8 months ago

Thank you very much, @kasumi-1, my bad. I will close the issue.

Is this documented anywhere?

The reason I thought it was LLaMA 2 is that it appears as the suggested default model if you open the Ollama settings:

image

kasumi-1 commented 8 months ago

I don't think it is documented, just from launch announcement. I'll add to the faq on heyamica.com - thank you

Aspie96 commented 8 months ago

Thank you a lot!

Please, consider updating Ollama default settings for consistency.