After updating GGUF, using lora via HOOKS causes an error: AttributeError: 'Linear' object has no attribute 'temp'

[rollingcookies]

At first KSampler started but there were a lot of errors in the terminal like:
ERROR lora diffusion_model.single_blocks.9.linear1.weight shape '[3072, 1728]' is invalid for input of size 9437184

Then, updated the GGUF and I get this:

!!! Exception during processing !!! 'Linear' object has no attribute 'temp'
Traceback (most recent call last):
File "D:!ComfyUI\ComfyUI\execution.py", line 323, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\execution.py", line 198, in get_output_data
return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\execution.py", line 169, in _map_node_over_list
process_inputs(input_dict, i)
File "D:!ComfyUI\ComfyUI\execution.py", line 158, in process_inputs
results.append(getattr(obj, func)(**inputs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\nodes.py", line 1465, in sample
return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\nodes.py", line 1432, in common_ksampler
samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\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 "D:!ComfyUI\ComfyUI\custom_nodes\ComfyUI-AnimateDiff-Evolved\animatediff\sampling.py", line 420, in motion_sample
return orig_comfy_sample(model, noise, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\sampling.py", line 116, in acn_sample
return orig_comfy_sample(model, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\utils.py", line 117, in uncond_multiplier_check_cn_sample
return orig_comfy_sample(model, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\sample.py", line 43, in sample
samples = sampler.sample(noise, positive, negative, 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 "D:!ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 117, in KSampler_sample
return orig_fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\custom_nodes\ComfyUI-TiledDiffusion\utils.py", line 51, in KSampler_sample
return orig_fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 1013, 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 "D:!ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 135, in sample
return orig_fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 911, in sample
return cfg_guider.sample(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 897, in sample
output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\patcher_extension.py", line 110, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 866, in outer_sample
output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 850, in inner_sample
samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\patcher_extension.py", line 110, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 100, in KSAMPLER_sample
return orig_fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\custom_nodes\ComfyUI-TiledDiffusion\utils.py", line 34, in KSAMPLER_sample
return orig_fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 707, in sample
samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\python_embeded\Lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\k_diffusion\sampling.py", line 155, in sample_euler
denoised = model(x, sigma_hat * s_in, **extra_args)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 379, in call
out = self.inner_model(x, sigma, model_options=model_options, seed=seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 832, in call
return self.predict_noise(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 835, in predict_noise
return sampling_function(self.inner_model, x, timestep, self.conds.get("negative", None), self.conds.get("positive", None), self.cfg, model_options=model_options, seed=seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 175, in sampling_function
out = orig_fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 359, in sampling_function
out = calc_cond_batch(model, conds, x, timestep, model_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 195, in calc_cond_batch
return executor.execute(model, conds, x_in, timestep, model_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\patcher_extension.py", line 110, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\samplers.py", line 277, in _calc_cond_batch
transformer_options = model.current_patcher.apply_hooks(hooks=hooks)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\model_patcher.py", line 1012, in apply_hooks
self.patch_hooks(hooks=hooks)
File "D:!ComfyUI\ComfyUI\comfy\model_patcher.py", line 1039, in patch_hooks
original_weights = self.get_key_patches()
^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\model_patcher.py", line 472, in get_key_patches
weight, set_func, convert_func = get_key_weight(self.model, k)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\model_patcher.py", line 121, in get_key_weight
op = comfy.utils.get_attr(model, op_keys[0])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\ComfyUI\comfy\utils.py", line 626, in get_attr
obj = getattr(obj, name)
^^^^^^^^^^^^^^^^^^
File "D:!ComfyUI\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1688, in getattr
raise AttributeError(f"'{type(self).name}' object has no attribute '{name}'")
AttributeError: 'Linear' object has no attribute 'temp'

[DevouredBeef]

Specifically stems from the changes added to comfyui in comfyanonymous/ComfyUI#5583

[alex-mitov]

I was just searching for this error. I get the same message with scheduled LoRAs and GGUF checkpoints.

[Tophness]

Specifically stems from the changes added to comfyui in comfyanonymous/ComfyUI#5583

I tried the ComfyUI commit just before that - comfyanonymous/ComfyUI/commit/2d5b3e0078c927ec6fcf47f80bf4035706934605 in combination with an old commit of GGUF (98333480059a2ccafb4718924ebcb9cdcb9b1f43) and it didn't work.
I went from getting the 'Linear' object has no attribute 'temp' error to hundreds of ERROR lora diffusion_model.single_blocks.9.linear1.weight shape '[3072, 1728]' is invalid for input of size 12345678 errors just like OP. The final output didn't use the LoRa in the region at all.

[DevouredBeef]

@Tophness the hook nodes don't exist prior to that PR being merged, there's no issues with regular lora loading currently, it's only when scheduled with hooks. If there were issues with regular lora loading (or something else) prior to the PR, it's clearly been corrected since then.

I suspect the nature of implementing GGUF is incompatible with the model patching changes surrounding the hooks in ComfyUI.

@city96 maybe you can confirm it's even feasible for the GGUF logic to be made compatible with the LoRa hooks?

[Jeremy-ttt]

Same problem when I use ModelMergerFlux1, any solution?

[city96]

@DevouredBeef Unsure, I haven't looked at how the hooks logic works so I can't comment on what it would take to implement. The LoRA logic for gguf quite different (and fragile) so I assume the issue is with that.

[Maelstrom2014]

same problem with GGUF models. 16vram two low to run with two lora models.