search

enricoros / big-AGI

Generative AI suite powered by state-of-the-art models and providing advanced AI/AGI functions. It features AI personas, AGI functions, multi-model chats, text-to-image, voice, response streaming, code highlighting and execution, PDF import, presets for developers, much more. Deploy on-prem or in the cloud.
https://big-agi.com
MIT License
4.9k stars 1.11k forks source link

Draw: Vendor: Add A1111 #368

Open Guerdal opened 5 months ago

Guerdal commented 5 months ago

Why Just like Big-AGI can interface with local LLM (for instance, ollama), it would be cool to have the ability in the Draw section to interface with Automatic1111 which possesses an API (link provided below). This would allow for independent operation of this feature. Description As for interfacing with Dall-E or Prodia, it would be necessary to add the capability to input a local URL of this type: http://IP_SERVER:7860/api

Requirements Interface Big-AGI with the API of AUTOMATIC1111 : https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/API

enricoros commented 5 months ago

Thanks @Guerdal, it's a good idea. What are the top projects that expose Local APIs for drawing? I'm looking to prioritize this effort, especially if there's some project that's really easy to install and ship alongside Big-AGI.

jtang613 commented 5 months ago

+1. The stable-diffusion-webui (aka Automatic1111) is by far the most popular locally-hosted text-to-image tool. It has a very clear API. It would be awesome to have this local option alongside Dall-E and Prodia.

enricoros commented 5 months ago

Thanks @jtang613 , added to the roadmap (unscheduled).

ACiDGRiM commented 4 months ago

I'd like to add that it would be great to also add localai as an api for stable diffusion and any of it's other compatible diffusers.