Make tile resample support SDXL model

[leonary]

Tile resample is a great feature, but it seems that there is no tile resample model that supports SDXL. Is it possible to use tile resample on SDXL?

[klgr]

Pretty dissapointed in wasn't working after updating to the new version. I got "mat1 and mat2 shapes cannot be multiplied" error when trying tile and inpaint

[philodoxos]

kohya has a blur controlnet, which can serve similar function to some extent...

[sdbds]

Now that it's on my calendar, I'll post it as soon as possible

[sdbds]

i think it should enough to use, i will upload it soon.

[camoody1]

@sdbds Excited for this! Please release as soon as you think it's ready.

[Diamond-Shark-art]

This looks really great, i've been waiting for Tile support for many weeks now so really happy its finally happening. Thank you!

[sdbds]

https://huggingface.co/bdsqlsz/qinglong_controlnet-lllite i had published tile_anime version for testing. photography version will come soon.

[ogasan]

Looking forward to it!

[YashamonSensei]

Hey, @sdbds Any updates on photo version? Sorry to bother you, but I'm waiting for this since I started using XL! Thanks!

[sdbds]

Hey, @sdbds Any updates on photo version? Sorry to bother you, but I'm waiting for this since I started using XL! Thanks!

It's actually finished, but I'm still finding time to test it. Because there's so much new stuff lately, LCM, SVD, SDXL-turbo ....

[camoody1]

@sdbds Oh boy!

[mickyboys]

Hey, @sdbds Any updates on photo version? Sorry to bother you, but I'm waiting for this since I started using XL! Thanks!

It's actually finished, but I'm still finding time to test it. Because there's so much new stuff lately, LCM, SVD, SDXL-turbo ....

Leave it up to the community to support and provide you with potential issues if existing. We are all desperately in need of your awesome tile version, so share it :-)

[cvar66]

Looking forward to the photo version!

[nasiri]

just rls it mate. !!! 🗡️

[techmd84]

Cant wait for the photo version! Ill test the beta!

[Momilijaz96]

I am down too.... we need this!!

[J-Cott]

photography version will come soon.

Is this Valve time soon? or AI time soon? :)

[camoody1]

"soon"

[sdbds]

I know most of you are looking forward and anxious, but hold your horses for now. Currently, because I'm not too satisfied with SDXL's photography base model, which doesn't work very well in terms of generalizability, I'm training a photographic base model using the A6000 ADA. When this model is trained, then the controlnet lllite tile will be trained. Please be patient as I've already spent $100 on electricity in the last month.

[Diamond-Shark-art]

Thanks for the update. Have you seen the latest JuggernautXL version, v8? It’s really good at photorealistic gens.

If you want some financial help, please drop a PayPal link or something like that. Would like to support your work.

[sdbds]

Thanks for the update. Have you seen the latest JuggernautXL version, v8? It’s really good at photorealistic gens.

If you want some financial help, please drop a PayPal link or something like that. Would like to support your work.

Thank you for respect！I just add a buymeacoffee link in my github page，if someone insteresed with my work can support with this way (also paypal with my email is OK) and I will add their credits on the model release page.

[Dervlex]

Any news on uploading the photo version ? Thx for your work!

[camoody1]

Knock, knock...

Anyone home? Controlnet Tile, sir? 🙏

[szriru]

Still waiting (´｡• ᵕ •｡`)

[camoody1]

@szriru I've given up hope. It's just not going to happen for SDXL. I figure the developer has sold the tiling to a company to earn to money.

[Dervlex]

Magnific 🫠🫠🫠

DarthBuckeye @.***> schrieb am Sa., 17. Feb. 2024, 22:48:

@szriru https://github.com/szriru I've given up hope. It's just not going to happen for SDXL. I figure the developer has sold the tiling to a company to earn to money.

— Reply to this email directly, view it on GitHub https://github.com/Mikubill/sd-webui-controlnet/issues/2049#issuecomment-1950392681, or unsubscribe https://github.com/notifications/unsubscribe-auth/BB753MGPMI4AJZFURJ3A2QLYUEQRBAVCNFSM6AAAAAA4LPHWWCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNJQGM4TENRYGE . You are receiving this because you commented.Message ID: @.***>

[huchenlei]

XAdapter's weight was just released. We should be able to use X-Adapter to apply SD15 tile ControlNet onto SDXL model. They actually has that example in their repo: https://github.com/showlab/X-Adapter/tree/main

[camoody1]

@Dervlex Bingo

[sdbds]

https://huggingface.co/bdsqlsz/qinglong_controlnet-lllite/ just come out~

[camoody1]

I can't seem to get this to work in ComfyUI. Getting this error message in the log:

error checkpoint does not contain controlnet or t2i adapter data H:\Stable_Diffusion_Models\models/ControlNet\ControlXL-Lora\bdsqlsz_controlllite_xl_tile_realistic.safetensors
ERROR:root:!!! Exception during processing !!!
ERROR:root:Traceback (most recent call last):
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 152, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 82, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 75, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\nodes.py", line 770, in apply_controlnet
    c_net = control_net.copy().set_cond_hint(control_hint, strength)
            ^^^^^^^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'copy'

[sdbds]

I can't seem to get this to work in ComfyUI. Getting this error message in the log:

error checkpoint does not contain controlnet or t2i adapter data H:\Stable_Diffusion_Models\models/ControlNet\ControlXL-Lora\bdsqlsz_controlllite_xl_tile_realistic.safetensors
ERROR:root:!!! Exception during processing !!!
ERROR:root:Traceback (most recent call last):
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 152, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 82, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 75, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\nodes.py", line 770, in apply_controlnet
    c_net = control_net.copy().set_cond_hint(control_hint, strength)
            ^^^^^^^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'copy'

comfyui use another node called controlllite not original controlnet

[camoody1]

@sdbds Is there any chance you might have a working .json you could share?

[Dervlex]

Those are my tries with automatic 1111 .....

Which settings are recommended ? sdbds ?

I dont know why, but it seems like i cant upscale it quite good with it.

[camoody1]

This IS supposed to be used with Ultimate SD Upscaler in ComfyUI, right? I know I used the ControlNet Tile 1.5 version in Automatic1111, but I never tried it in ComfyUI. Maybe I'm just trying to use it completely incorrectly. I'm using a node called "Load LLLite", now, but I'm still getting an error:

ERROR:root:!!! Exception during processing !!!
ERROR:root:Traceback (most recent call last):
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 152, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 82, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 75, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale\nodes.py", line 125, in upscale
    processed = script.run(p=sdprocessing, _=None, tile_width=tile_width, tile_height=tile_height,
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale\repositories\ultimate_sd_upscale\scripts\ultimate-upscale.py", line 553, in run
    upscaler.process()
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale\repositories\ultimate_sd_upscale\scripts\ultimate-upscale.py", line 136, in process
    self.image = self.redraw.start(self.p, self.image, self.rows, self.cols)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale\repositories\ultimate_sd_upscale\scripts\ultimate-upscale.py", line 243, in start
    return self.linear_process(p, image, rows, cols)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale\repositories\ultimate_sd_upscale\scripts\ultimate-upscale.py", line 178, in linear_process
    processed = processing.process_images(p)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale\modules\processing.py", line 122, in process_images
    (samples,) = common_ksampler(p.model, p.seed, p.steps, p.cfg, p.sampler_name,
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\nodes.py", line 1338, in common_ksampler
    samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image,
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\sample_error_enhancer.py", line 22, in informative_sample
    raise e
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\sample_error_enhancer.py", line 9, in informative_sample
    return original_sample(*args, **kwargs)  # This code helps interpret error messages that occur within exceptions but does not have any impact on other operations.
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AnimateDiff-Evolved\animatediff\sampling.py", line 248, in motion_sample
    return orig_comfy_sample(model, noise, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\sample.py", line 100, in sample
    samples = sampler.sample(noise, positive_copy, negative_copy, cfg=cfg, latent_image=latent_image, start_step=start_step, last_step=last_step, force_full_denoise=force_full_denoise, denoise_mask=noise_mask, sigmas=sigmas, callback=callback, disable_pbar=disable_pbar, seed=seed)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_smZNodes\__init__.py", line 130, in KSampler_sample
    return _KSampler_sample(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\samplers.py", line 706, in sample
    return sample(self.model, noise, positive, negative, cfg, self.device, sampler, sigmas, self.model_options, latent_image=latent_image, denoise_mask=denoise_mask, callback=callback, disable_pbar=disable_pbar, seed=seed)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_smZNodes\__init__.py", line 149, in sample
    return _sample(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\samplers.py", line 611, in sample
    samples = sampler.sample(model_wrap, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\samplers.py", line 550, in sample
    samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\k_diffusion\sampling.py", line 580, in sample_dpmpp_2m
    denoised = model(x, sigmas[i] * s_in, **extra_args)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\samplers.py", line 282, in forward
    out = self.inner_model(x, sigma, cond=cond, uncond=uncond, cond_scale=cond_scale, model_options=model_options, seed=seed)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\samplers.py", line 272, in forward
    return self.apply_model(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 1012, in apply_model
    out = super().apply_model(*args, **kwargs)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\samplers.py", line 269, in apply_model
    out = sampling_function(self.inner_model, x, timestep, uncond, cond, cond_scale, model_options=model_options, seed=seed)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\samplers.py", line 249, in sampling_function
    cond_pred, uncond_pred = calc_cond_uncond_batch(model, cond, uncond_, x, timestep, model_options)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-TiledDiffusion\.patches.py", line 4, in calc_cond_uncond_batch
    return calc_cond_uncond_batch_original_tiled_diffusion_87234190(model, cond, uncond, x_in, timestep, model_options)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\samplers.py", line 223, in calc_cond_uncond_batch
    output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\model_base.py", line 91, in apply_model
    model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds).float()
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL\modules\custom_sdxl_ksampler.py", line 70, in new_unet_forward
    x0 = old_unet_forward(self, x, timesteps, context, y, control, transformer_options, **kwargs)
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\diffusionmodules\openaimodel.py", line 847, in forward
    h = forward_timestep_embed(module, h, emb, context, transformer_options, time_context=time_context, num_video_frames=num_video_frames, image_only_indicator=image_only_indicator)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\diffusionmodules\openaimodel.py", line 43, in forward_timestep_embed
    x = layer(x, context, transformer_options)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\attention.py", line 632, in forward
    x = block(x, context=context[i], transformer_options=transformer_options)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\attention.py", line 459, in forward
    return checkpoint(self._forward, (x, context, transformer_options), self.parameters(), self.checkpoint)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\diffusionmodules\util.py", line 189, in checkpoint
    return func(*inputs)
           ^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\ldm\modules\attention.py", line 498, in _forward
    n, context_attn1, value_attn1 = p(n, context_attn1, value_attn1, extra_options)
                                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ControlNet-LLLite-ComfyUI\node_control_net_lllite.py", line 111, in __call__
    q = q + self.modules[module_pfx_to_q](q)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ControlNet-LLLite-ComfyUI\node_control_net_lllite.py", line 243, in forward
    cx = torch.cat([cx, self.down(x)], dim=1 if self.is_conv2d else 2)
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: Sizes of tensors must match except in dimension 2. Expected size 4096 but got size 4356 for tensor number 1 in the list.

Prompt executed in 11.82 seconds

[DarkStoorM]

If you get mismatched tensor types error, disable Hypertile. At least on A1111 that works for me.

[camoody1]

@DarkStoorM That's not an option that I'm aware of in Comfy. So, it's working well for you in Automatic1111? I'm probably just an idiot and don't know how to workflow SHOULD work in Comfy.

[DarkStoorM]

@DarkStoorM That's not an option that I'm aware of in Comfy. So, it's working well for you in Automatic1111? I'm probably just an idiot and don't know how to workflow SHOULD work in Comfy.

Never used Comfy, so can't tell what's the solution there. Someone has to post a workflow if they get it to work so you get a node comparison, there must be something missing. or a config at least.

I'd suggest using it on a very minimal workflow to test it, but judging by this error it might be Tiled Diffusion at fault, but can't really tell.

[xpeng]

https://huggingface.co/bdsqlsz/qinglong_controlnet-lllite/ just come out~

good result! how to set in Automatic1111 controlnet? just sellect model to xl_tile_realistic like this?

[sdbds]

@sdbds Is there any chance you might have a working .json you could share?

try use this node to load controlllite

https://github.com/yolain/ComfyUI-Easy-Use

[sdbds]

https://huggingface.co/bdsqlsz/qinglong_controlnet-lllite/ just come out~

good result! how to set in Automatic1111 controlnet? just sellect model to xl_tile_realistic like this?

just use as SD1.5

[altoiddealer]

This IS supposed to be used with Ultimate SD Upscaler in ComfyUI, right? I know I used the ControlNet Tile 1.5 version in Automatic1111, but I never tried it in ComfyUI. Maybe I'm just trying to use it completely incorrectly. I'm using a node called "Load LLLite", now, but I'm still getting an error:

Unsure how much has changed since then, but tiled upscaling was always most recommended by ControlNet team for use with the Tiled VAE/Diffusion extension (AKA Multidiffusion). This extension is built-in and recommended with sdwebui-forge (by lllyasviel the author of Controlnet).

https://github.com/Mikubill/sd-webui-controlnet

Now ControlNet 1.1 can support almost all Upscaling/Tile methods. ControlNet 1.1 support the script "Ultimate SD upscale" and almost all other tile-based extensions. Please do not confuse "Ultimate SD upscale" with "SD upscale" - they are different scripts. Note that the most recommended upscaling method is "Tiled VAE/Diffusion" but we test as many methods/extensions as possible. Note that "SD upscale" is supported since 1.1.117, and if you use it, you need to leave all ControlNet images as blank (We do not recommend "SD upscale" since it is somewhat buggy and cannot be maintained - use the "Ultimate SD upscale" instead).

[dotsimulate]

seems to be working in comfyui with https://github.com/Kosinkadink/ComfyUI-Advanced-ControlNet

[camoody1]

seems to be working in comfyui with https://github.com/Kosinkadink/ComfyUI-Advanced-ControlNet

Hey, you're right! I WAS able to get this to work. Sadly, it doesn't seem to work nearly as well as the SD1.5 version. With a high enough denoise, the result looks very, very clean with this new Tile CN, but it doesn't add the incredible detail like the SD1.5 version was so well known for.

[sdbds]

Hey, you're right! I WAS able to get this to work. Sadly, it doesn't seem to work nearly as well as the SD1.5 version. With a high enough denoise, the result looks very, very clean with this new Tile CN, but it doesn't add the incredible detail like the SD1.5 version was so well known for.

we recommend use tile diffusion to add noise within denoising process

[EricRollei]

There's a full SDXL controlnet tile available on CivitAi here... https://civitai.com/models/330313/tplanetsdxlcontrolnettilerealisticv1 have been testing this morning. Looks pretty good. I downloaded the lllite realistic tile, but can't get it to work with the lllite loader node?