Closed dfl closed 5 months ago
PS I also had a dependency error with the latest diffusers, which requires huggingface-hub==0.20.3 (instead of 0.20.1)
one more issue...
Installing llama-cpp-python...
ERROR: llama_cpp_python-0.2.43-macosx_14_0_arm64.whl is not a valid wheel filename.
Traceback (most recent call last):
File "/Users/dfl/sd/ComfyUI/nodes.py", line 1893, in load_custom_node
module_spec.loader.exec_module(module)
File "<frozen importlib._bootstrap_external>", line 940, in exec_module
File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
File "/Users/dfl/sd/ComfyUI/custom_nodes/ComfyUI_VLM_nodes/__init__.py", line 30, in <module>
install_llama(system_info)
File "/Users/dfl/sd/ComfyUI/custom_nodes/ComfyUI_VLM_nodes/install_init.py", line 93, in install_llama
install_package("llama-cpp-python", custom_command=custom_command)
File "/Users/dfl/sd/ComfyUI/custom_nodes/ComfyUI_VLM_nodes/install_init.py", line 73, in install_package
subprocess.check_call(command)
File "/opt/homebrew/Cellar/python@3.11/3.11.7_1/Frameworks/Python.framework/Versions/3.11/lib/python3.11/subprocess.py", line 413, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['/Users/dfl/sd/ComfyUI/venv/bin/python', '-m', 'pip', 'install', 'llama-cpp-python', '--no-cache-dir', 'https://github.com/abetlen/llama-cpp-python/releases/download/v0.2.43/llama_cpp_python-0.2.43-macosx_14_0_arm64.whl']' returned non-zero exit status 1.
Cannot import /Users/dfl/sd/ComfyUI/custom_nodes/ComfyUI_VLM_nodes module for custom nodes: Command '['/Users/dfl/sd/ComfyUI/venv/bin/python', '-m', 'pip', 'install', 'llama-cpp-python', '--no-cache-dir', 'https://github.com/abetlen/llama-cpp-python/releases/download/v0.2.43/llama_cpp_python-0.2.43-macosx_14_0_arm64.whl']' returned non-zero exit status 1.
installing manually worked:
(base) ➜ ComfyUI git:(master) ✗ ./venv/bin/pip install llama-cpp-python
Looking in indexes: https://pypi.org/simple, https://pypi.ngc.nvidia.com
Collecting llama-cpp-python
Downloading llama_cpp_python-0.2.43.tar.gz (36.6 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 36.6/36.6 MB 14.8 MB/s eta 0:00:00
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing metadata (pyproject.toml) ... done
Requirement already satisfied: typing-extensions>=4.5.0 in ./venv/lib/python3.11/site-packages (from llama-cpp-python) (4.8.0)
Requirement already satisfied: numpy>=1.20.0 in ./venv/lib/python3.11/site-packages (from llama-cpp-python) (1.24.4)
Requirement already satisfied: diskcache>=5.6.1 in ./venv/lib/python3.11/site-packages (from llama-cpp-python) (5.6.3)
Requirement already satisfied: jinja2>=2.11.3 in ./venv/lib/python3.11/site-packages (from llama-cpp-python) (3.1.2)
Requirement already satisfied: MarkupSafe>=2.0 in ./venv/lib/python3.11/site-packages (from jinja2>=2.11.3->llama-cpp-python) (2.1.3)
Building wheels for collected packages: llama-cpp-python
Building wheel for llama-cpp-python (pyproject.toml) ... done
Created wheel for llama-cpp-python: filename=llama_cpp_python-0.2.43-cp311-cp311-macosx_14_0_arm64.whl size=2265033 sha256=afb33ffff14641c2d0da077601821e547400e2b6957060a56727987a9339708e
Stored in directory: /private/var/folders/71/7jzsfc7x6xx8t1fyzxphfhgh0000gn/T/pip-ephem-wheel-cache-zw0tsmjq/wheels/2a/9d/08/b558eac0caff83868235db10ed76b5b1d2ec06276cdd26dc5d
Successfully built llama-cpp-python
Installing collected packages: llama-cpp-python
Successfully installed llama-cpp-python-0.2.43
and finally: AssertionError: Torch not compiled with CUDA enabled
is it working after this changes
yes, after changing CUDA stuff it works 👍🏼On 15 Feb 2024, at 12:28, gokayfem @.***> wrote: is it working after this changes
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: @.***>
mac m1max ,This node cannot be installed properly. How to deal with it?
autogptq library does not support macOS. so i didnt implement other things suggested in this issue for macOS also. i might need to create another branch only for mac. i dont own a mac, i cant quickly iterate on that sorry.
mac m1max ,This node cannot be installed properly. How to deal with it?
i created another branch called "mac" and implemented the suggested things in this issue. if you can try to download the mac branch and put in your custom_nodes folder you can use it i think. if you get any error please let me now in this issue.
mac m1max ,This node cannot be installed properly. How to deal with it?
i created another branch called "mac" and implemented the suggested things in this issue. if you can try to download the mac branch and put in your custom_nodes folder you can use it i think. if you get any error please let me now in this issue.
Great, but I just tried to install it and it still won't import!
can you provide me the errors please. i dont have a mac, i cant check it.
can you provide me the errors please. i dont have a mac, i cant check it.我没有 Mac,无法检查。
Here is the error message from the terminal: Traceback (most recent call last): File "/Users/weiwei/ComfyUI/nodes.py", line 1887, in load_custom_node module_spec.loader.exec_module(module) File "", line 883, in exec_module File "", line 241, in _call_with_frames_removed File "/Users/weiwei/ComfyUI/custom_nodes/ComfyUI_VLM_nodes/init.py", line 29, in system_info = get_system_info() File "/Users/weiwei/ComfyUI/custom_nodes/ComfyUI_VLM_nodes/install_init.py", line 52, in get_system_info system_info['avx2'] = 'avx2' in cpuinfo.get_cpu_info()['flags'] KeyError: 'flags'
Cannot import /Users/weiwei/ComfyUI/custom_nodes/ComfyUI_VLM_nodes module for custom nodes: 'flags'
1-) First go to the mac branch
2-) Then download the zip
3-) Delete original VLM nodes
4-)Extract the zip then try to start comfyui again.
Thanks so much, the install worked with the zip!
hello, thanks for all your work on this project!
I'm running on a Mac M3, and get_cpu_info() dict does not contain 'flags' key.
This makes it work: