File "F:\SD\sd-webui-aki-v4.6.1\modules\processing.py", line 1257, in sample
samples = self.sampler.sample(self, x, conditioning, unconditional_conditioning, image_conditioning=self.txt2img_image_conditioning(x))
File "F:\SD\sd-webui-aki-v4.6.1\modules\sd_samplers_kdiffusion.py", line 234, in sample
samples = self.launch_sampling(steps, lambda: self.func(self.model_wrap_cfg, x, extra_args=self.sampler_extra_args, disable=False, callback=self.callback_state, **extra_params_kwargs))
File "F:\SD\sd-webui-aki-v4.6.1\modules\sd_samplers_common.py", line 261, in launch_sampling
return func()
File "F:\SD\sd-webui-aki-v4.6.1\modules\sd_samplers_kdiffusion.py", line 234, in <lambda>
samples = self.launch_sampling(steps, lambda: self.func(self.model_wrap_cfg, x, extra_args=self.sampler_extra_args, disable=False, callback=self.callback_state, **extra_params_kwargs))
File "F:\SD\sd-webui-aki-v4.6.1\python\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "F:\SD\sd-webui-aki-v4.6.1\repositories\k-diffusion\k_diffusion\sampling.py", line 128, in sample_euler
denoised = model(x, sigma_hat * s_in, **extra_args)
File "F:\SD\sd-webui-aki-v4.6.1\python\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "F:\SD\sd-webui-aki-v4.6.1\python\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "F:\SD\sd-webui-aki-v4.6.1\modules\sd_samplers_cfg_denoiser.py", line 269, in forward
devices.test_for_nans(x_out, "unet")
File "F:\SD\sd-webui-aki-v4.6.1\modules\devices.py", line 255, in test_for_nans
raise NansException(message)
modules.devices.NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check.
---

有没有懂的老哥

那段英文最后不是教你如何设置了么。。。

同

提高精度：在模型的设置中，将 “Upcast cross attention layer to float32” 选项勾选，这样可以将模型的某些层从半精度浮点数（float16）提升到全精度浮点数（float32），以提高计算精度。
命令行参数：使用命令行参数 --no-half 来运行模型，这样可以禁用半精度浮点数的使用，确保所有计算都在全精度下进行。
禁用NaN检查：作为临时解决方案，您可以使用 --disable-nan-check 命令行参数来禁用NaN检查。但请注意，这可能会掩盖潜在的问题，不是长久之计。

4g的显卡，放弃挣扎，放弃治疗。显存不够。