oobabooga on another server - how to setup?

[LightTemplar]

How to setup, if I have oobabooga installed on another machine?

[if-ai]

I will add a way to dynamically set it on the settings but for now the only way is by manually change the ip on the scrips/if_prompt_mkr.py

[if-ai]

I added the feature to use another ip check the video later is uploading now

[LightTemplar]

From the logs looks like connection is working! Though, I still didn't get that beautiful result from your images :)

[if-ai]

what is the prompt you get? did you get the prompt, could you take a screenshot of the image + prompt?

[LightTemplar]

Thanks to your video, I got it working =)
I only had to use TheBloke/WizardLM-7B-V1-0-Uncensored-SuperHOT-8K-GGML with Transformers because I ain't got enough vram for 13B GPTQ model with Exllama.
My result for the promt "(CatGirl warrior:1.2), legendary flower" was:

[if-ai]

That's fantastic thank you so much for showing your result. Did it work with server or only local. I will make my own checks this weekend but it would be nice to know, so far I could only run local, so is hard for me to tell if something doesn't work if people don't tell me.

…

On Tue, Aug 8, 2023, 6:35 AM Light Templar ***@***.***> wrote: Closed #3 <#3> as completed. — Reply to this email directly, view it on GitHub <#3 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFBUFQRXBVGO6WQ3FFPTY43XUHF3TANCNFSM6AAAAAA2XSOLTU> . You are receiving this because you commented.Message ID: ***@***.***>

[LightTemplar]

I understand! They are two VMs on two PCs in local network.
For oobabooga I got 5GB VRAM (Quadro P2200), and for AUTO1111 - 12 GB VRAM (RTX 3060)