-
-
hope support lsm https://huggingface.co/latent-consistency
-
### Issue Description
When I use any sd 1.5 model I have with lcm's sd 1.5 lora and set the sampler to LCM as I must , the results are garbled if I enable HYPERTILE.
I have to use hypertile becaus…
-
The Inference speed of those model is almost instant,
Adding a small image to webpages could add to the immersion with a small cost in time to generate, can this be done? maybe as an option?
-
A list of ideas to explore:
* [x] Lazy transfers (so we don't load data into the GPU at once)
* [x] FP16 on load
* [x] FP16 policies on Axon
* [x] ~[Attention slicing](https://github.com/hugging…
-
Hi how I can change the denoising strength with vid2vid ? I thinks it's to high by default, thanks
-
Hi there. The LCM sambler is the "new" goodness in A1111. Image generation can be done much faster and they come out better looking than using the standard Euler sampler.
And new custom loras and mod…
CHJ85 updated
6 months ago
-
Hey you can add non lcm diffusers to lcm mode I tried openjourney but the image was cloudy. We need the maxed cfg guidance scale for lcm mode please add cfg higher scale at least cfg 0.1-10.0 please m…
-
Are there any plans to support LoRA?
If so, I assume the `.safetensors` file will need to be converted with the model?
-
On [this build](https://github.com/leejet/stable-diffusion.cpp/releases/tag/master-760cfaa) most of the lora's i have will cause a black image as an output.
I'm not entirely sure why this would be th…