Vladimir Sotnikov

Any updates on this one?

@patrickvonplaten unfortunately, your code sample results in the same bug: ``` Traceback (most recent call last): File "/home/vladimir/miniconda3/envs/SD/lib/python3.9/code.py", line 90, in runcode exec(code, self.locals) File "", line 12, in File...

+1, this negatively affects my performance as well :( The workaround I'm currently using is to provide an explicit type hint upon an instantiation of a pipeline: 

Use `from ffmpeg import probe` or ```python import ffmpeg probe = ffmpeg.probe(...) ```

Are you sure you need FFmpeg for that? You could read bytes directly from the file: ```python PACKET_SIZE = 100 # number of bytes, could be any number with open("video.mp4",...

`check_repository_consistency` failed so I added this fix to AltDiffusion as well.

Sure! Should I make an fp16 version of the test as well, or fp32 only would be enough?

ONNX img2img pipeline is actually failing when I try to use image that is divisible by 8 but not 16 or 32: ``` ╭───────────────────── Traceback (most recent call last) ──────────────────────╮...

Actually, it looks like ONNX pipeline can't even work with resolutions that are multiples of 32, only 64 are supported. This code uses the init image that is a multiple...

Could we for now apply this fix only to `StableDiffusionImg2ImgPipeline` and `AltDiffusionImg2ImgPipeline`, and keep the ONNX pipeline intact? :) Given that there's already a discrepancy in text2img between these three...