Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Unable to pin memory: operation not supported #49

Open
booksword opened this issue Dec 10, 2024 · 5 comments
Open

Unable to pin memory: operation not supported #49

booksword opened this issue Dec 10, 2024 · 5 comments

Comments

@booksword
Copy link

(venv) ubuntu@hp:/nunchaku$ ls
app assets build comfyui example.py LICENCE.txt MANIFEST.in nunchaku nunchaku.egg-info pyproject.toml README.md schnell.model setup.py src third_party
(venv) ubuntu@hp:
/nunchaku$ python example.py
[2024-12-10 10:42:02.110] [info] Initializing QuantizedFluxModel
[2024-12-10 10:42:02.243] [info] Loading weights from /home/ubuntu/.cache/huggingface/hub/models--mit-han-lab--svdq-int4-flux.1-schnell/snapshots/3957d6ece9852f09ccadc064f1fd25964394a490/transformer_blocks.safetensors
[2024-12-10 10:42:02.243] [warning] Unable to pin memory: operation not supported
[2024-12-10 10:42:02.243] [info] Try MAP_PRIVATE
Traceback (most recent call last):
File "/home/ubuntu/nunchaku/example.py", line 5, in
transformer = NunchakuFluxTransformer2dModel.from_pretrained("mit-han-lab/svdq-int4-flux.1-schnell")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/ubuntu/ComfyUI/venv/lib/python3.12/site-packages/huggingface_hub/utils/_validators.py", line 114, in _inner_fn
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/home/ubuntu/nunchaku/nunchaku/models/transformer_flux.py", line 206, in from_pretrained
m = load_quantized_module(transformer_block_path, device=device)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/ubuntu/nunchaku/nunchaku/models/transformer_flux.py", line 114, in load_quantized_module
m.load(path)
RuntimeError: CUDA error: invalid argument (at /home/ubuntu/nunchaku/src/Serialization.cpp:101)
(venv) ubuntu@hp:~/nunchaku$ nvidia-smi
Tue Dec 10 10:42:14 2024
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 565.57.01 Driver Version: 565.57.01 CUDA Version: 12.7 |
|-----------------------------------------+------------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA GeForce RTX 4060 Off | 00000000:01:00.0 Off | N/A |
| 33% 32C P8 N/A / 115W | 2MiB / 8188MiB | 0% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| No running processes found |
+-----------------------------------------------------------------------------------------+
(venv) ubuntu@hp:~/nunchaku$

@booksword
Copy link
Author

not support 4060 :(

@lmxyy
Copy link
Collaborator

lmxyy commented Dec 10, 2024

4060 was supported. However, since it only has 8G GPU memory, it may require some offloading.

@wardensc2
Copy link

wardensc2 commented Dec 13, 2024

Hi @lmxyy My 3090 with 24gb Vram get this warning too and crash ComfyUI after this. I have download full model but can't run

image

@iiiian
Copy link

iiiian commented Dec 19, 2024

my 4060ti 16GB is having the same issue.

@vbqgyy
Copy link

vbqgyy commented Dec 19, 2024

WIN10 3090 64gRAM comfyui having the same issue. torch2.4.1+cu124 ,already installed cuda12.6

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants