ComfyUI
ComfyUI copied to clipboard
Published
20 hours ago •
comfyanonymous
comfyanonymous
FP8 not running claiming I need ROCm 6.5 or higher, but I believe I am running up to ROCm 7.x
trafficstars
Custom Node Testing
- [x] I have tried disabling custom nodes and the issue persists (see how to disable custom nodes if you need help)
Expected Behavior
Trying to generate a Wan 2.2.1 text to video using:
Actual Behavior
It errors out when hitting the KSampler(Advanced) node.
I believe the issue originates with:
Steps to Reproduce
Running latest comfy with latest interface - the template has not been changed and all models are correct.
Debug Logs
got prompt
Requested to load WanTEModel
loaded completely; 8282.29 MB usable, 6419.48 MB loaded, full load: True
0 models unloaded.
loaded partially; 6315.36 MB usable, 6306.08 MB loaded, 7321.43 MB offloaded, lowvram patches: 0
0%| | 0/2 [00:00<?, ?it/s]
!!! Exception during processing !!! FP8 _scaled_mm failed, falling back to dequantization: Float8_e4m3fn is only supported for ROCm 6.5 and above
Traceback (most recent call last):
File "/home/mark/Apps/ComfyUI/comfy/quant_ops.py", line 446, in fp8_linear
output = torch._scaled_mm(
plain_input.reshape(-1, input_shape[2]).contiguous(),
...<4 lines>...
out_dtype=out_dtype,
)
RuntimeError: Float8_e4m3fn is only supported for ROCm 6.5 and above
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/mark/Apps/ComfyUI/execution.py", line 510, in execute
output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/execution.py", line 324, in get_output_data
return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/execution.py", line 298, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "/home/mark/Apps/ComfyUI/execution.py", line 286, in process_inputs
result = f(**inputs)
File "/home/mark/Apps/ComfyUI/nodes.py", line 1559, in sample
return common_ksampler(model, noise_seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise, disable_noise=disable_noise, start_step=start_at_step, last_step=end_at_step, force_full_denoise=force_full_denoise)
File "/home/mark/Apps/ComfyUI/nodes.py", line 1492, in common_ksampler
samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image,
denoise=denoise, disable_noise=disable_noise, start_step=start_step, last_step=last_step,
force_full_denoise=force_full_denoise, noise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=seed)
File "/home/mark/Apps/ComfyUI/comfy/sample.py", line 60, 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 "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 1163, 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 "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 1053, in sample
return cfg_guider.sample(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 1035, in sample
output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)
File "/home/mark/Apps/ComfyUI/comfy/patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 997, in outer_sample
output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 980, in inner_sample
samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)
File "/home/mark/Apps/ComfyUI/comfy/patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 752, in sample
samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)
File "/home/mark/Apps/ComfyUI/venv/lib/python3.13/site-packages/torch/utils/_contextlib.py", line 120, in decorate_context
return func(*args, **kwargs)
File "/home/mark/Apps/ComfyUI/comfy/k_diffusion/sampling.py", line 199, in sample_euler
denoised = model(x, sigma_hat * s_in, **extra_args)
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 401, in __call__
out = self.inner_model(x, sigma, model_options=model_options, seed=seed)
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 953, in __call__
return self.outer_predict_noise(*args, **kwargs)
~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 960, in outer_predict_noise
).execute(x, timestep, model_options, seed)
~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 963, 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 "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 381, in sampling_function
out = calc_cond_batch(model, conds, x, timestep, model_options)
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 206, in calc_cond_batch
return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 214, in _calc_cond_batch_outer
return executor.execute(model, conds, x_in, timestep, model_options)
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/samplers.py", line 326, in _calc_cond_batch
output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/model_base.py", line 161, in apply_model
return comfy.patcher_extension.WrapperExecutor.new_class_executor(
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
...<2 lines>...
comfy.patcher_extension.get_all_wrappers(comfy.patcher_extension.WrappersMP.APPLY_MODEL, transformer_options)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
).execute(x, t, c_concat, c_crossattn, control, transformer_options, **kwargs)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/model_base.py", line 203, in _apply_model
model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)
File "/home/mark/Apps/ComfyUI/venv/lib/python3.13/site-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/venv/lib/python3.13/site-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
File "/home/mark/Apps/ComfyUI/comfy/ldm/wan/model.py", line 627, in forward
return comfy.patcher_extension.WrapperExecutor.new_class_executor(
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
...<2 lines>...
comfy.patcher_extension.get_all_wrappers(comfy.patcher_extension.WrappersMP.DIFFUSION_MODEL, transformer_options)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
).execute(x, timestep, context, clip_fea, time_dim_concat, transformer_options, **kwargs)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/patcher_extension.py", line 112, in execute
return self.original(*args, **kwargs)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/ldm/wan/model.py", line 647, in _forward
return self.forward_orig(x, timestep, context, clip_fea=clip_fea, freqs=freqs, transformer_options=transformer_options, **kwargs)[:, :, :t, :h, :w]
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/ldm/wan/model.py", line 547, in forward_orig
e = self.time_embedding(
sinusoidal_embedding_1d(self.freq_dim, t.flatten()).to(dtype=x[0].dtype))
File "/home/mark/Apps/ComfyUI/venv/lib/python3.13/site-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/venv/lib/python3.13/site-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
File "/home/mark/Apps/ComfyUI/venv/lib/python3.13/site-packages/torch/nn/modules/container.py", line 250, in forward
input = module(input)
File "/home/mark/Apps/ComfyUI/venv/lib/python3.13/site-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/venv/lib/python3.13/site-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
File "/home/mark/Apps/ComfyUI/comfy/ops.py", line 161, in forward
return self.forward_comfy_cast_weights(*args, **kwargs)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/home/mark/Apps/ComfyUI/comfy/ops.py", line 486, in forward_comfy_cast_weights
out = fp8_linear(self, input)
File "/home/mark/Apps/ComfyUI/comfy/ops.py", line 436, in fp8_linear
o = torch.nn.functional.linear(quantized_input, quantized_weight, bias)
File "/home/mark/Apps/ComfyUI/comfy/quant_ops.py", line 211, in __torch_dispatch__
return handler(func, args, kwargs)
File "/home/mark/Apps/ComfyUI/comfy/quant_ops.py", line 472, in fp8_linear
raise RuntimeError(f"FP8 _scaled_mm failed, falling back to dequantization: {e}")
RuntimeError: FP8 _scaled_mm failed, falling back to dequantization: Float8_e4m3fn is only supported for ROCm 6.5 and above
Prompt executed in 2.60 seconds
Other
Im running CachyOS Linux, AMD RX 9070XT GPU.
