SamplerCustomAdvanced Bias is not supported when out_dtype is set to Float32

[bjsongzhen]

ComfyUI Error Report

Error Details

• Node Type: SamplerCustomAdvanced
• Exception Type: RuntimeError

• Exception Message: Bias is not supported when out_dtype is set to Float32

  Stack Trace

  
  File "E:\ComfyUI-aki-v1.4F1\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 "E:\ComfyUI-aki-v1.4F1\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 "E:\ComfyUI-aki-v1.4F1\execution.py", line 169, in _map_node_over_list
  process_inputs(input_dict, i)
  
  File "E:\ComfyUI-aki-v1.4F1\execution.py", line 158, in process_inputs
  results.append(getattr(obj, func)(**inputs))
  
  File "E:\ComfyUI-aki-v1.4F1\comfy_extras\nodes_custom_sampler.py", line 633, in sample
  samples = guider.sample(noise.generate_noise(latent), latent_image, sampler, sigmas, denoise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=noise.seed)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\samplers.py", line 716, in sample
  output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\samplers.py", line 695, in inner_sample
  samples = sampler.sample(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)
  
  File "E:\ComfyUI-aki-v1.4F1\custom_nodes\ComfyUI-TiledDiffusion\utils.py", line 34, in KSAMPLER_sample
  return orig_fn(*args, **kwargs)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\samplers.py", line 600, in sample
  samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)
  
  File "E:\ComfyUI-aki-v1.4F1\python\lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context
  return func(*args, **kwargs)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\k_diffusion\sampling.py", line 155, in sample_euler
  denoised = model(x, sigma_hat * s_in, **extra_args)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\samplers.py", line 299, in __call__
  out = self.inner_model(x, sigma, model_options=model_options, seed=seed)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\samplers.py", line 682, in __call__
  return self.predict_noise(*args, **kwargs)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\samplers.py", line 685, 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 "E:\ComfyUI-aki-v1.4F1\comfy\samplers.py", line 279, in sampling_function
  out = calc_cond_batch(model, conds, x, timestep, model_options)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\samplers.py", line 228, in calc_cond_batch
  output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)
  
  File "E:\ComfyUI-aki-v1.4F1\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\utils.py", line 69, in apply_model_uncond_cleanup_wrapper
  return orig_apply_model(self, *args, **kwargs)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\model_base.py", line 142, in apply_model
  model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds).float()
  
  File "E:\ComfyUI-aki-v1.4F1\python\lib\site-packages\torch\nn\modules\module.py", line 1553, in _wrapped_call_impl
  return self._call_impl(*args, **kwargs)
  
  File "E:\ComfyUI-aki-v1.4F1\python\lib\site-packages\torch\nn\modules\module.py", line 1562, in _call_impl
  return forward_call(*args, **kwargs)
  
  File "E:\ComfyUI-aki-v1.4F1\custom_nodes\ComfyUI-Fluxtapoz\flux\model.py", line 137, in forward
  out = self.forward_orig(img, img_ids_orig, context, txt_ids, timestep, y, guidance, control, transformer_options=transformer_options, ref_img_ids=ref_img_ids)
  
  File "E:\ComfyUI-aki-v1.4F1\custom_nodes\ComfyUI-Fluxtapoz\flux\model.py", line 33, in forward_orig
  img = self.img_in(img)
  
  File "E:\ComfyUI-aki-v1.4F1\python\lib\site-packages\torch\nn\modules\module.py", line 1553, in _wrapped_call_impl
  return self._call_impl(*args, **kwargs)
  
  File "E:\ComfyUI-aki-v1.4F1\python\lib\site-packages\torch\nn\modules\module.py", line 1562, in _call_impl
  return forward_call(*args, **kwargs)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\ops.py", line 67, in forward
  return self.forward_comfy_cast_weights(*args, **kwargs)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\ops.py", line 285, in forward_comfy_cast_weights
  out = fp8_linear(self, input)
  
  File "E:\ComfyUI-aki-v1.4F1\comfy\ops.py", line 267, in fp8_linear
  o = torch._scaled_mm(inn, w, out_dtype=input.dtype, bias=bias, scale_a=scale_input, scale_b=scale_weight)

## System Information
- **ComfyUI Version:** v0.2.3-19-g73e3a9e6
- **Arguments:** E:\ComfyUI-aki-v1.4F1\main.py --auto-launch --preview-method auto --disable-cuda-malloc --fast
- **OS:** nt
- **Python Version:** 3.10.11 (tags/v3.10.11:7d4cc5a, Apr  5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)]
- **Embedded Python:** false
- **PyTorch Version:** 2.4.1+cu124
## Devices

- **Name:** cuda:0 NVIDIA GeForce RTX 4090 : cudaMallocAsync
  - **Type:** cuda
  - **VRAM Total:** 25756696576
  - **VRAM Free:** 6242935036
  - **Torch VRAM Total:** 17817403392
  - **Torch VRAM Free:** 673108220

[bjsongzhen]

When I was using the example_flux_reference.json and reflux_workflow_example_updated.json workflow, an error occurred. but other workflows are working normally.

[tianshanxin]

同样问题

[tianshanxin]

大佬解决下

[bjsongzhen]

This seems to be a problem with "configure ref for flux" node

[zhilengjun]

UNETLoader weight_dtype choose dafult well work

[bjsongzhen]

UNETLoader weight_dtype choose dafult well work

ComfyUI Error Report

Error Details

• Node Type: SamplerCustomAdvanced
• Exception Type: torch.OutOfMemoryError
• Exception Message: Allocation on device

  Stack Trace

After I made the changes, a new error occurred.

[logtd]

Hi, thanks for testing this out.

Unfortunately the Ref Sampling features you're testing are not completed. I plan to revisit these soon. If you're looking for image editing you can use the rf-inversion workflow

https://github.com/logtd/ComfyUI-Fluxtapoz/blob/main/example_workflows/example_rf_inversion_updated.json

[zhilengjun]

UNETLoader weight_dtype choose dafult well work

ComfyUI Error Report

Error Details

• Node Type: SamplerCustomAdvanced
• Exception Type: torch.OutOfMemoryError
• Exception Message: Allocation on device

Stack Trace

After I made the changes, a new error occurred.

Maybe there is insufficient VRAM.Try setting a smaller image size.

[bjsongzhen]

Hi, thanks for testing this out.

Unfortunately the Ref Sampling features you're testing are not completed. I plan to revisit these soon. If you're looking for image editing you can use the rf-inversion workflow

https://github.com/logtd/ComfyUI-Fluxtapoz/blob/main/example_workflows/example_rf_inversion_updated.json

thinks,its work!