Invoke is a leading creative engine for Stable Diffusion models, empowering professionals, artists, and enthusiasts to generate and create visual media using the latest AI-driven technologies. The solution offers an industry leading WebUI, and serves as the foundation for multiple commercial products.
Integrating Ella with InvokeAI would allow us to write more detailed prompts, which are followed more closely by SD.
The only unknown I have is if Compel (which is using OpenCLIP) would work well with ELLA (which is using a t5 LLM instead of CLIP), and what approach might be taken here to improve prompt adherence while not giving up on the advanced prompting syntax.
Is there an existing issue for this?
Contact Details
No response
What should this feature add?
Integrating Ella with InvokeAI would allow us to write more detailed prompts, which are followed more closely by SD.
The only unknown I have is if Compel (which is using OpenCLIP) would work well with ELLA (which is using a t5 LLM instead of CLIP), and what approach might be taken here to improve prompt adherence while not giving up on the advanced prompting syntax.
Alternatives
No response
Additional Content
No response