Unload to CPU: VideoAutoencoderKL
Load to GPU: AutoencoderKL
Unload to CPU: AutoencoderKL
Load to GPU: CLIPTextModel
Unload to CPU: CLIPTextModel
Load to GPU: ModifiedUNet
Unload to CPU: ModifiedUNet
Load to GPU: AutoencoderKL
Unload to CPU: AutoencoderKL
Load to GPU: CLIPTextModel
Unload to CPU: CLIPTextModel
Load to GPU: Resampler
Load to GPU: ImprovedCLIPVisionModelWithProjection
Unload to CPU: Resampler
Unload to CPU: ImprovedCLIPVisionModelWithProjection
Load to GPU: VideoAutoencoderKL
Traceback (most recent call last):
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\queueing.py", line 528, in process_events
response = await route_utils.call_process_api(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\route_utils.py", line 270, in call_process_api
output = await app.get_blocks().process_api(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\blocks.py", line 1908, in process_api
result = await self.call_function(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\blocks.py", line 1485, in call_function
prediction = await anyio.to_thread.run_sync(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio\to_thread.py", line 56, in run_sync
return await get_async_backend().run_sync_in_worker_thread(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio_backends_asyncio.py", line 2177, in run_sync_in_worker_thread
return await future
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio_backends_asyncio.py", line 859, in run
result = context.run(func, args)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\utils.py", line 808, in wrapper
response = f(args, kwargs)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, *kwargs)
File "E:\UNDO\Paints-UNDO\gradio_app.py", line 222, in process_video
frames, im1, im2 = process_video_inner(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(args, kwargs)
File "E:\UNDO\Paints-UNDO\gradio_app.py", line 190, in process_video_inner
input_frame_latents, vae_hidden_states = video_pipe.encode_latents(input_frames, return_hidden_states=True)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, kwargs)
File "E:\UNDO\Paints-UNDO\diffusers_vdm\pipeline.py", line 102, in encode_latents
encoder_posterior, hidden_states = self.vae.encode(x, return_hidden_states=return_hidden_states)
File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 804, in encode
h, hidden = self.encoder(x, return_hidden_states)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, *kwargs)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl
return forward_call(args, kwargs)
File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 249, in forward
h = self.mid.attn_1(h)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, kwargs)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl
return forward_call(*args, kwargs)
File "E:\UNDO\Paints-UNDO\diffusersvdm\vae.py", line 411, in forward
h = self.attention(h_)
File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 397, in attention
out = chunked_attention(
File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 36, in chunked_attention
out = xformers.ops.memory_efficient_attention(q, k, v)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha__init.py", line 276, in memory_efficient_attention
return _memory_efficient_attention(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\init.py", line 395, in _memory_efficient_attention
return _memory_efficient_attention_forward(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\init__.py", line 414, in _memory_efficient_attention_forward
op = _dispatch_fw(inp, False)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\dispatch.py", line 119, in _dispatch_fw
return _run_priority_list(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\dispatch.py", line 55, in _run_priority_list
raise NotImplementedError(msg)
NotImplementedError: No operator found for memory_efficient_attention_forward with inputs:
query : shape=(2, 2688, 1, 512) (torch.float16)
key : shape=(2, 2688, 1, 512) (torch.float16)
value : shape=(2, 2688, 1, 512) (torch.float16)
attn_bias : <class 'NoneType'>
p : 0.0
decoderF is not supported because:
max(query.shape[-1] != value.shape[-1]) > 128
xFormers wasn't build with CUDA support
attn_bias type is <class 'NoneType'>
operator wasn't built - see python -m xformers.info for more info
flshattF@0.0.0 is not supported because:
max(query.shape[-1] != value.shape[-1]) > 256
xFormers wasn't build with CUDA support
operator wasn't built - see python -m xformers.info for more info
cutlassF is not supported because:
xFormers wasn't build with CUDA support
operator wasn't built - see python -m xformers.info for more info
smallkF is not supported because:
max(query.shape[-1] != value.shape[-1]) > 32
xFormers wasn't build with CUDA support
dtype=torch.float16 (supported: {torch.float32})
operator wasn't built - see python -m xformers.info for more info
unsupported embed per head: 512
Unload to CPU: VideoAutoencoderKL
Load to GPU: CLIPTextModel
Unload to CPU: CLIPTextModel
Load to GPU: Resampler
Load to GPU: ImprovedCLIPVisionModelWithProjection
Unload to CPU: Resampler
Unload to CPU: ImprovedCLIPVisionModelWithProjection
Load to GPU: VideoAutoencoderKL
Traceback (most recent call last):
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\queueing.py", line 528, in process_events
response = await route_utils.call_process_api(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\route_utils.py", line 270, in call_process_api
output = await app.get_blocks().process_api(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\blocks.py", line 1908, in process_api
result = await self.call_function(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\blocks.py", line 1485, in call_function
prediction = await anyio.to_thread.run_sync(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio\to_thread.py", line 56, in run_sync
return await get_async_backend().run_sync_in_worker_thread(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio_backends_asyncio.py", line 2177, in run_sync_in_worker_thread
return await future
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio_backends_asyncio.py", line 859, in run
result = context.run(func, args)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\utils.py", line 808, in wrapper
response = f(args, kwargs)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, kwargs)
File "E:\UNDO\Paints-UNDO\gradio_app.py", line 222, in process_video
frames, im1, im2 = process_video_inner(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, kwargs)
File "E:\UNDO\Paints-UNDO\gradio_app.py", line 190, in process_video_inner
input_frame_latents, vae_hidden_states = video_pipe.encode_latents(input_frames, return_hidden_states=True)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, *kwargs)
File "E:\UNDO\Paints-UNDO\diffusers_vdm\pipeline.py", line 102, in encode_latents
encoder_posterior, hidden_states = self.vae.encode(x, return_hidden_states=return_hidden_states)
File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 804, in encode
h, hidden = self.encoder(x, return_hidden_states)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl
return self._call_impl(args, kwargs)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl
return forward_call(*args, kwargs)
File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 249, in forward
h = self.mid.attn_1(h)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, *kwargs)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl
return forward_call(args, kwargs)
File "E:\UNDO\Paints-UNDO\diffusersvdm\vae.py", line 411, in forward
h = self.attention(h_)
File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 397, in attention
out = chunked_attention(
File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 36, in chunked_attention
out = xformers.ops.memory_efficient_attention(q, k, v)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha__init.py", line 276, in memory_efficient_attention
return _memory_efficient_attention(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\init.py", line 395, in _memory_efficient_attention
return _memory_efficient_attention_forward(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\init__.py", line 414, in _memory_efficient_attention_forward
op = _dispatch_fw(inp, False)
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\dispatch.py", line 119, in _dispatch_fw
return _run_priority_list(
File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\dispatch.py", line 55, in _run_priority_list
raise NotImplementedError(msg)
NotImplementedError: No operator found for memory_efficient_attention_forward with inputs:
query : shape=(2, 2688, 1, 512) (torch.float16)
key : shape=(2, 2688, 1, 512) (torch.float16)
value : shape=(2, 2688, 1, 512) (torch.float16)
attn_bias : <class 'NoneType'>
p : 0.0
decoderF is not supported because:
max(query.shape[-1] != value.shape[-1]) > 128
xFormers wasn't build with CUDA support
attn_bias type is <class 'NoneType'>
operator wasn't built - see python -m xformers.info for more info
flshattF@0.0.0 is not supported because:
max(query.shape[-1] != value.shape[-1]) > 256
xFormers wasn't build with CUDA support
operator wasn't built - see python -m xformers.info for more info
cutlassF is not supported because:
xFormers wasn't build with CUDA support
operator wasn't built - see python -m xformers.info for more info
smallkF is not supported because:
max(query.shape[-1] != value.shape[-1]) > 32
xFormers wasn't build with CUDA support
dtype=torch.float16 (supported: {torch.float32})
operator wasn't built - see python -m xformers.info for more info
unsupported embed per head: 512
Windows 11
13th Gen Intel(R) Core(TM) i7-13700K 3.40 GHz
RTX 4090
Unload to CPU: VideoAutoencoderKL Load to GPU: AutoencoderKL Unload to CPU: AutoencoderKL Load to GPU: CLIPTextModel Unload to CPU: CLIPTextModel Load to GPU: ModifiedUNet Unload to CPU: ModifiedUNet Load to GPU: AutoencoderKL Unload to CPU: AutoencoderKL Load to GPU: CLIPTextModel Unload to CPU: CLIPTextModel Load to GPU: Resampler Load to GPU: ImprovedCLIPVisionModelWithProjection Unload to CPU: Resampler Unload to CPU: ImprovedCLIPVisionModelWithProjection Load to GPU: VideoAutoencoderKL Traceback (most recent call last): File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\queueing.py", line 528, in process_events response = await route_utils.call_process_api( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\route_utils.py", line 270, in call_process_api output = await app.get_blocks().process_api( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\blocks.py", line 1908, in process_api result = await self.call_function( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\blocks.py", line 1485, in call_function prediction = await anyio.to_thread.run_sync( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio\to_thread.py", line 56, in run_sync return await get_async_backend().run_sync_in_worker_thread( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio_backends_asyncio.py", line 2177, in run_sync_in_worker_thread return await future File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio_backends_asyncio.py", line 859, in run result = context.run(func, args) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\utils.py", line 808, in wrapper response = f(args, kwargs) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context return func(*args, *kwargs) File "E:\UNDO\Paints-UNDO\gradio_app.py", line 222, in process_video frames, im1, im2 = process_video_inner( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context return func(args, kwargs) File "E:\UNDO\Paints-UNDO\gradio_app.py", line 190, in process_video_inner input_frame_latents, vae_hidden_states = video_pipe.encode_latents(input_frames, return_hidden_states=True) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context return func(*args, kwargs) File "E:\UNDO\Paints-UNDO\diffusers_vdm\pipeline.py", line 102, in encode_latents encoder_posterior, hidden_states = self.vae.encode(x, return_hidden_states=return_hidden_states) File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 804, in encode h, hidden = self.encoder(x, return_hidden_states) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl return self._call_impl(*args, *kwargs) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl return forward_call(args, kwargs) File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 249, in forward h = self.mid.attn_1(h) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl return self._call_impl(*args, kwargs) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl return forward_call(*args, kwargs) File "E:\UNDO\Paints-UNDO\diffusersvdm\vae.py", line 411, in forward h = self.attention(h_) File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 397, in attention out = chunked_attention( File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 36, in chunked_attention out = xformers.ops.memory_efficient_attention(q, k, v) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha__init.py", line 276, in memory_efficient_attention return _memory_efficient_attention( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\init.py", line 395, in _memory_efficient_attention return _memory_efficient_attention_forward( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\init__.py", line 414, in _memory_efficient_attention_forward op = _dispatch_fw(inp, False) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\dispatch.py", line 119, in _dispatch_fw return _run_priority_list( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\dispatch.py", line 55, in _run_priority_list raise NotImplementedError(msg) NotImplementedError: No operator found for
memory_efficient_attention_forward
with inputs: query : shape=(2, 2688, 1, 512) (torch.float16) key : shape=(2, 2688, 1, 512) (torch.float16) value : shape=(2, 2688, 1, 512) (torch.float16) attn_bias : <class 'NoneType'> p : 0.0decoderF
is not supported because: max(query.shape[-1] != value.shape[-1]) > 128 xFormers wasn't build with CUDA support attn_bias type is <class 'NoneType'> operator wasn't built - seepython -m xformers.info
for more infoflshattF@0.0.0
is not supported because: max(query.shape[-1] != value.shape[-1]) > 256 xFormers wasn't build with CUDA support operator wasn't built - seepython -m xformers.info
for more infocutlassF
is not supported because: xFormers wasn't build with CUDA support operator wasn't built - seepython -m xformers.info
for more infosmallkF
is not supported because: max(query.shape[-1] != value.shape[-1]) > 32 xFormers wasn't build with CUDA support dtype=torch.float16 (supported: {torch.float32}) operator wasn't built - seepython -m xformers.info
for more info unsupported embed per head: 512 Unload to CPU: VideoAutoencoderKL Load to GPU: CLIPTextModel Unload to CPU: CLIPTextModel Load to GPU: Resampler Load to GPU: ImprovedCLIPVisionModelWithProjection Unload to CPU: Resampler Unload to CPU: ImprovedCLIPVisionModelWithProjection Load to GPU: VideoAutoencoderKL Traceback (most recent call last): File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\queueing.py", line 528, in process_events response = await route_utils.call_process_api( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\route_utils.py", line 270, in call_process_api output = await app.get_blocks().process_api( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\blocks.py", line 1908, in process_api result = await self.call_function( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\blocks.py", line 1485, in call_function prediction = await anyio.to_thread.run_sync( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio\to_thread.py", line 56, in run_sync return await get_async_backend().run_sync_in_worker_thread( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio_backends_asyncio.py", line 2177, in run_sync_in_worker_thread return await future File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\anyio_backends_asyncio.py", line 859, in run result = context.run(func, args) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\gradio\utils.py", line 808, in wrapper response = f(args, kwargs) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context return func(*args, kwargs) File "E:\UNDO\Paints-UNDO\gradio_app.py", line 222, in process_video frames, im1, im2 = process_video_inner( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context return func(*args, kwargs) File "E:\UNDO\Paints-UNDO\gradio_app.py", line 190, in process_video_inner input_frame_latents, vae_hidden_states = video_pipe.encode_latents(input_frames, return_hidden_states=True) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context return func(*args, *kwargs) File "E:\UNDO\Paints-UNDO\diffusers_vdm\pipeline.py", line 102, in encode_latents encoder_posterior, hidden_states = self.vae.encode(x, return_hidden_states=return_hidden_states) File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 804, in encode h, hidden = self.encoder(x, return_hidden_states) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl return self._call_impl(args, kwargs) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl return forward_call(*args, kwargs) File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 249, in forward h = self.mid.attn_1(h) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl return self._call_impl(*args, *kwargs) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl return forward_call(args, kwargs) File "E:\UNDO\Paints-UNDO\diffusersvdm\vae.py", line 411, in forward h = self.attention(h_) File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 397, in attention out = chunked_attention( File "E:\UNDO\Paints-UNDO\diffusers_vdm\vae.py", line 36, in chunked_attention out = xformers.ops.memory_efficient_attention(q, k, v) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha__init.py", line 276, in memory_efficient_attention return _memory_efficient_attention( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\init.py", line 395, in _memory_efficient_attention return _memory_efficient_attention_forward( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\init__.py", line 414, in _memory_efficient_attention_forward op = _dispatch_fw(inp, False) File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\dispatch.py", line 119, in _dispatch_fw return _run_priority_list( File "C:\Users\DM.conda\envs\paints_undo\lib\site-packages\xformers\ops\fmha\dispatch.py", line 55, in _run_priority_list raise NotImplementedError(msg) NotImplementedError: No operator found formemory_efficient_attention_forward
with inputs: query : shape=(2, 2688, 1, 512) (torch.float16) key : shape=(2, 2688, 1, 512) (torch.float16) value : shape=(2, 2688, 1, 512) (torch.float16) attn_bias : <class 'NoneType'> p : 0.0decoderF
is not supported because: max(query.shape[-1] != value.shape[-1]) > 128 xFormers wasn't build with CUDA support attn_bias type is <class 'NoneType'> operator wasn't built - seepython -m xformers.info
for more infoflshattF@0.0.0
is not supported because: max(query.shape[-1] != value.shape[-1]) > 256 xFormers wasn't build with CUDA support operator wasn't built - seepython -m xformers.info
for more infocutlassF
is not supported because: xFormers wasn't build with CUDA support operator wasn't built - seepython -m xformers.info
for more infosmallkF
is not supported because: max(query.shape[-1] != value.shape[-1]) > 32 xFormers wasn't build with CUDA support dtype=torch.float16 (supported: {torch.float32}) operator wasn't built - seepython -m xformers.info
for more info unsupported embed per head: 51213th Gen Intel(R) Core(TM) i7-13700K 3.40 GHz RTX 4090