You can support me here : https://ko-fi.com/photobait if you enjoy my 600+ free Loras (including the update) of fictional characters, if you want to encourage me and help me create more Loras. Or just reward me with Buzz.
To request a specific/custom Flux1.D Lora using buzz:
- 2500 Buzz if the Lora can be shared publicly (SFW),
- 5000 Buzz if the Lora have to remain private.
For SDXL Lora, please contact me directly.
Description
FAQ
Comments (3)
Do you use A1111/Forge or ComfyUI to generate your Flux images? I've been using Forge with some trained models and q5+ks or w/e it's called and I keep getting garbage results. I try it with DEIS and Beta sampler off of some recommendations, but Flux never seems to listen to my prompt well, takes up more VRAM than any chart says it should (probably from Loras), and takes longer than SDXL/Pony by far to make results that ignore half my prompts and look half as good. Is it a Forge thing? Is the sampler wrong? I'd love to be able to use your models but I can't seem to get Flux to produce good results.
I use forge too and very happy with it. I'll share my config to you tomorrow
@Phraxas , as promised : https://civitai.com/articles/9349
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.