Invoke AI Latest Version Windows, RunPod and Massed Compute 1-Click Installers

Furkan Gözükara

Thanks Alex exactly like that.

Alex

did you follow the steps in the read me? i also have a 5090 and i ran the install then moved the 5000 bat to the root of the invoke folder its working flawless for me

Furkan Gözükara

you are making mistake somewhere. if you upgrade gold tier i can connect your pc and fix it.

Furkan Gözükara

it is related to your computer. you if logout from your hugging face via cli it should get fixed

Kornél Garamvölgyi

And i got cuda error when i start any task. RuntimeError: CUDA error: no kernel image is available for execution on the device CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1 Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. I have an 5090 card. I run the upgrade rtx 5000 bat. But no success.

Kornél Garamvölgyi

Is there any solution to prevent this error message: "Invalid or Missing HF Token" and allow the models to be downloaded? This issue only occurs with some models (Flux 1 redux dev, Fulx.1 depth dev lora, Flux.1 canny, Stable Diffusion 3.5 large).

Furkan Gözükara

Updated and now there is upgrade for RTX 5000 series

Furkan Gözükara

Updated and now there is upgrade for RTX 5000 series

Furkan Gözükara

Gonna try to make it work now

Furkan Gözükara

Gonna try to make it work now

Alex

anyway you can update this for 50 series cards?

Hassan Alhassan

5090?

Furkan Gözükara

thank you so much

renderartist

Thank you for this, nice work!

Christopher Guillou

Thanks ! May just end up trying Runpod...

Furkan Gözükara

let me test. on runpod i tested it worked

Christopher Guillou

Massed Compute : Oh well... seems we're still getting "./invoke.sh: line 37: 8333 Segmentation fault (core dumped) invokeai-web $PARAMS"

Furkan Gözükara

hocam v3 ile fixledim teşekkürler bilgilendirme için. lütfen v3 zip dosyasını yükle

Furkan Gözükara

hocam v3 ile fixledim teşekkürler bilgilendirme için. lütfen v3 zip dosyasını yükle

Bayram TATAR

Evet harfiyyen aynısını uyguladım

Furkan Gözükara

hocam bu videoyu adım adım izleyip hepsini kurdunuz mu? https://youtu.be/DrhUHnYfwC0

Furkan Gözükara

hocam bu videoyu adım adım izleyip hepsini kurdunuz mu? https://youtu.be/DrhUHnYfwC0

Bayram TATAR

Bayram TATAR

Furkan Gözükara

i would say you have upload error to the runpod. how you are uploading / downloading on runpod?

Michael 'Kay' Klindt

When trying to install my Loras on Runpod 90% fail with runpod ValueError: Error while deserializing header: MetadataIncompleteBuffer. Do you know the issue? Some of the same exact Loras ran fine in Invoke locally, but I have been trying for days to get them to work in Runpod.

Furkan Gözükara

şu videoyu adım adım izleyip hepsini birebir yapabilir misin? https://youtu.be/DrhUHnYfwC0

Furkan Gözükara

yes install error. you need python 3.10, cuda 12.4, cudnn 8.9.7 and c++ tools. and also git. here entire tutorial for all : https://youtu.be/DrhUHnYfwC0

Royal Sero

hata veriyor furkan abi ne denedim olmadi, acemiyim biraz

Royal Sero

C:\Users\Serhat-Pc>python -c "import torch; print(torch.__version__)" 2.4.1+cu121

Royal Sero

C:\Users\Serhat-Pc>python Python 3.10.11 (tags/v3.10.11:7d4cc5a, Apr 5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> import torch >>> x = torch.rand(5, 3) >>> print(x) tensor([[0.8943, 0.1868, 0.1155], [0.3184, 0.5475, 0.1129], [0.4480, 0.2118, 0.7054], [0.3152, 0.1228, 0.9532], [0.1583, 0.1623, 0.5797]]) >>> >>> import torch >>> torch.cuda.is_available() True >>>

Royal Sero

Collecting bitsandbytes==0.43.3 (from invokeai[onnx-cuda,xformers]) Using cached bitsandbytes-0.43.3-py3-none-win_amd64.whl.metadata (3.5 kB) Collecting windows-curses (from invokeai[onnx-cuda,xformers]) Using cached windows_curses-2.3.3-cp310-cp310-win_amd64.whl.metadata (2.7 kB) Collecting xformers==0.0.28.post1 (from invokeai[onnx-cuda,xformers]) Using cached xformers-0.0.28.post1.tar.gz (7.8 MB) Installing build dependencies ... done Getting requirements to build wheel ... error error: subprocess-exited-with-error × Getting requirements to build wheel did not run successfully. │ exit code: 1 ╰─> [17 lines of output] Traceback (most recent call last): File "B:\xInvokeAI\.venv\lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 353, in main() File "B:\xInvokeAI\.venv\lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 335, in main json_out['return_val'] = hook(**hook_input['kwargs']) File "B:\xInvokeAI\.venv\lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 118, in get_requires_for_build_wheel return hook(config_settings) File "C:\Users\Serhat-Pc\AppData\Local\Temp\pip-build-env-d8e8p5a_\overlay\Lib\site-packages\setuptools\build_meta.py", line 332, in get_requires_for_build_wheel return self._get_build_requires(config_settings, requirements=[]) File "C:\Users\Serhat-Pc\AppData\Local\Temp\pip-build-env-d8e8p5a_\overlay\Lib\site-packages\setuptools\build_meta.py", line 302, in _get_build_requires self.run_setup() File "C:\Users\Serhat-Pc\AppData\Local\Temp\pip-build-env-d8e8p5a_\overlay\Lib\site-packages\setuptools\build_meta.py", line 503, in run_setup super().run_setup(setup_script=setup_script) File "C:\Users\Serhat-Pc\AppData\Local\Temp\pip-build-env-d8e8p5a_\overlay\Lib\site-packages\setuptools\build_meta.py", line 318, in run_setup exec(code, locals()) File "", line 24, in ModuleNotFoundError: No module named 'torch' [end of output] note: This error originates from a subprocess, and is likely not a problem with pip. error: subprocess-exited-with-error × Getting requirements to build wheel did not run successfully. │ exit code: 1 ╰─> See above for output. note: This error originates from a subprocess, and is likely not a problem with pip. Error: Unexpected exit code: 1 Command line: | 'B:\xInvokeAI\.venv\Scripts\pip.exe' install --require-virtualenv --force-reinstall --use-pep517 'invokeai[xformers,onnx-cuda]' --extra-index-url https://download.pytorch.org/whl/cu121 Could not install InvokeAI. Please try downloading the latest version of the installer and install again. Drücken Sie eine beliebige Taste . . .

Furkan Gözükara

yes it has a linux system with desktop. it is better at applications that requires interface not web based. main advantage of massed compute is, a6000 gpu is so cheap and their harddrive and internet speed are amazing

C

So if I'm looking at this correctly, Massed compute creates a sort of Linux desktop? If so, this is much better to use than Runpod with just Jupyter lab, isn't it?

Furkan Gözükara

you are welcome. i just forwarded your message to them as well - first message because it causes confusion

Chris

Wow...that is one way to make money! Thanks for the clarification.

Furkan Gözükara

you have to terminate every time stop is useless. stop will use your money same as running. i told this to their team

Chris

I have a question about Massed Compute. Last time I stopped the machine, it started itself again after some time. I realized it after a few hours, so many credits where used without using the machine. It should be enough to just press "Stop" on the machine? I do not have to terminate it evey time i stop using it?

Furkan Gözükara

this is another UI that some people prefer. it is known by its inpainting is superior. i am recording a quick tutorial right now

Steve

this replace anything for anyone or another UI that does the same thing?