cog-stable-diffusion

cog-stable-diffusion copied to clipboard

Hi, Why is it that I am getting significantly different results from my local SD V1.5 using the same seed? How can I use the Euler A Sampler with this API?

3

ade5791

ValueError: operands could not be broadcast together with shapes (2,) (8,)

3

Version `a9758cb` this morning started to give me the following error for all img2img queries via the API: ``` Running predict()... Using seed: 36432201 Traceback (most recent call last): File...

matthewtoast

Is there a specific reason why you still use the legacy inpainting pipeline and not the new one? Why don’t you use the for inpainting optimized inpainting model?

blueturtleai

Support for selecting samplers

1

The user should be able to select between: * k-lms * pndm * ddim * euler_a * dpm_2_a

andreasjansson

To catch stuff like #40...

bfirsh

As suggested by @uglyrobot: https://github.com/replicate/cog-stable-diffusion/issues/25#issuecomment-1282979832 Maybe we detect how much RAM the connected GPU has and do the sensible thing. That way it'll work on any GPUs users are running...

bfirsh

From Discord: https://discord.com/channels/775512803439280149/890687279805370419/1020395081867464804 If you use a better noise generator than random noise for the masked regions, SD can be used for outpainting. As suggested by __luke__#1234, the noise generator...

andreasjansson

Increase the upper bound for guidance scale to 50.

andreasjansson

Couldn't find a solution to this error message online, but didn't want to post into the void of Discord in case someone else has this problem in the future. Command...

dginovker

How can we continuously push this model to Replicate?

andreasjansson