It is ok for use to generate images for the first time. However, it will raise the following error if we generate images for second time.

Traceback (most recent call last):
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/gradio/routes.py", line 321, in run_predict
    output = await app.blocks.process_api(
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/gradio/blocks.py", line 1006, in process_api
    result = await self.call_function(fn_index, inputs, iterator, request)
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/gradio/blocks.py", line 847, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/anyio/to_thread.py", line 31, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/anyio/_backends/_asyncio.py", line 937, in run_sync_in_worker_thread
    return await future
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/anyio/_backends/_asyncio.py", line 867, in run
    result = context.run(func, *args)
  File "app.py", line 123, in infer
    images = pipe_refiner(prompt=prompt, negative_prompt=negative, image=images, num_inference_steps=steps, strength=refiner_strength, generator=g).images
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl_img2img.py", line 998, in __call__
    image = self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False)[0]
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/diffusers/utils/accelerate_utils.py", line 46, in wrapper
    return method(self, *args, **kwargs)
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/diffusers/models/autoencoder_kl.py", line 270, in decode
    decoded = self._decode(z).sample
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/diffusers/models/autoencoder_kl.py", line 256, in _decode
    z = self.post_quant_conv(z)
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 463, in forward
    return self._conv_forward(input, self.weight, self.bias)
  File "/docker/software/anaconda3/envs/r3d/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 459, in _conv_forward
    return F.conv2d(input, weight, bias, self.stride,
RuntimeError: Input type (c10::Half) and bias type (float) should be the same

Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas->gradio==3.14.0->-r requirements.txt (line 11)) (1.16.0)
Requirement already satisfied: exceptiongroup in /usr/local/lib/python3.10/dist-packages (from anyio->httpx->gradio==3.14.0->-r requirements.txt (line 11)) (1.2.0)
Installing collected packages: triton, torch
Attempting uninstall: triton
Found existing installation: triton 2.1.0
Uninstalling triton-2.1.0:
Successfully uninstalled triton-2.1.0
Attempting uninstall: torch
Found existing installation: torch 2.1.0
Uninstalling torch-2.1.0:
Successfully uninstalled torch-2.1.0
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
torchaudio 2.1.0 requires torch==2.1.0, but you have torch 2.0.1 which is incompatible.
torchdata 0.7.0 requires torch==2.1.0, but you have torch 2.0.1 which is incompatible.
torchtext 0.16.0 requires torch==2.1.0, but you have torch 2.0.1 which is incompatible.
torchvision 0.16.0 requires torch==2.1.0, but you have torch 2.0.1 which is incompatible.
Successfully installed torch-2.0.1 triton-2.0.0

Is it possible to set specific seed for each image?

Hi,
I'm not a very experienced developer, so I can't seem to figure this out.
When I run your code in Google Colab I get only a local URL:

Running on local URL:  http://127.0.0.1:7860/
*** Failed to connect to ec2.gradio.app:22: [Errno 110] Connection timed out

I tried to make a fork in which I forced the share property to be true :

block.queue().launch(share="true")

But the problem persists. Any idea if I'm doing something wrong?

Thank you,
Erwin Velthuis

I am reading the SDXL paper and found that the refiner is applying to the latent image

but in your codes,

images = pipe_refiner(prompt=prompt, negative_prompt=negative, image=images, num_inference_steps=steps, strength=refiner_strength).images

the input are the images instead of the latents
are they the same?

thanks

colab dropdown, just turn it to html

 # @param ["1:1", "4:1", "16:9", "5:2", "2:1", "7:4", "3:2", "8:7", "9:8", "8:9", "7:8", "2:3", "4:7", "1:2", "2:5", "1:3", "9:16"] {allow-input: true}

python code

aspect_ratio = "9:16"
max_pixel = 1024*1024
w, h = (int(e) for e in aspect_ratio.split(':'))
width, height = (round(math.sqrt(max_pixel * x / y) / 8) * 8 for x, y in ((w, h), (h, w)))

width and height can be the param of the param of the model

I am quite surprised by the image quality of the SD XL 1.0 model, and in addition to the Refiner model, which according to my tests really improves the overall image quality.

I noticed that there is a little bit of GPU memory to generate images with a bit more resolution than 1024px. Could you please add the option to select aspect ratio and resolution in the google colab demo?

Next I'll show you my examples with and without Refiner. All generation parameters are completely identical, they are distinguished by Refiner on and Refiner off.

with Refiner

without Refiner

As we can see, this example alone is enough to see that Refiner really improves the image. And it is not bad, but it requires a lot of memory.

Thanks for your attention, and for your work!

Enabling Multi-GPU Support for SDXL

Dear developer,

I am currently using the SDXL for my project, and I am encountering some difficulties with enabling multi-GPU support. I have four Nvidia 3090 GPUs at my disposal, but so far, I have only been able to run the software on one of them, running such a large model on a single 3090 is very slow and ram cosuming.

My attempts to distribute the workload among all GPUs have been unsuccessful. These have included:

• Attempt : Use the accelerate to config the multi GPUS and run.

Unfortunately, these methods have not resulted in successful multi-GPU utilization.

For your reference, here is some information about my setup:

• Operating System: Ubuntu
• Python Version: 3.10.12
• CUDA Version:12.2
• Deep Learning Framework: PyTorch 2.0.1

I'm not sure if I'm missing something or if there's an issue with SDXL itself. Therefore, I'm writing to ask if you could provide some guidance on this matter. Is there a built-in way to enable multi-GPU support, or are there any additional steps that I might have overlooked? Thank you!