is ai town launguge model must be using openai's GPT api?
jimmy8399 opened this issue · comments
is ai town launguge model must be using openai's GPT api? or its option ?
There's no other option at the moment. Feel free to add another integration.
What if I want to test a local language model, such as LLaMA? how should I revise the code?
We're working on that. It's not supported yet.
…On Sat, Sep 2, 2023 at 9:30 AM chatsci ***@***.***> wrote:
What if I want to test a local language model, such as LLaMA? how should I
revise the code?
—
Reply to this email directly, view it on GitHub
<#133 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACWEFGHNE73DAIP3HV6UPTTXYNNK5ANCNFSM6AAAAAA4D4ARAM>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
What are all the models folks would like to see?
- LLaMA on Replicate
- LLaMA locally
Those are the top requests I've seen - any other interesting ones? And are folks looking for the llama-2-70b-chat or something else?
want to use llama-2-70b-chat locally
In #184 we add Ollama support which you can run locally for chat (but not any embeddings yet)
It landed! try it out and let us know how it's working for you. documented in readme