Intel Arc Graphics Thread #476
Replies: 51 comments 138 replies
-
|
UPDATE:
Before Intel update it, the code needs to be modified like this to avoid reporting errors: - match operation:
- case "multiply":
- output[top:bottom, left:right] = destination_portion * source_portion
- case "add":
- output[top:bottom, left:right] = destination_portion + source_portion
- case "subtract":
- output[top:bottom, left:right] = destination_portion - source_portion
+ if operation == "multiply":
+ output[top:bottom, left:right] = destination_portion * source_portion
+ elif operation == "add":
+ output[top:bottom, left:right] = destination_portion + source_portion
+ elif operation == "subtract":
+ output[top:bottom, left:right] = destination_portion - source_portionOf course not modifying it doesn't seem to affect normal usage. |
Beta Was this translation helpful? Give feedback.
-
|
I added this fix to master: 476d543 |
Beta Was this translation helpful? Give feedback.
-
|
About Karras scheduler: Currently using it causes a -996 "uses-fp64-math" error, which may be fixed in the next IPEX release (intel/intel-extension-for-pytorch#285) Until then, can specify to use # comfy/samplers.py
- sigmas = k_diffusion_sampling.get_sigmas_karras(n=steps, sigma_min=self.sigma_min, sigma_max=self.sigma_max, device=self.device)
+ sigmas = k_diffusion_sampling.get_sigmas_karras(n=steps, sigma_min=self.sigma_min, sigma_max=self.sigma_max, device="cpu") |
Beta Was this translation helpful? Give feedback.
-
|
is it Linux only or can you use this on a win10 machine? |
Beta Was this translation helpful? Give feedback.
-
|
You should be able to use WSL, but I haven't tested it. |
Beta Was this translation helpful? Give feedback.
-
|
So again, sharing the good news. An XPU version of their Pytorch extension with Pytorch 2.0 support,
Hats off to ComfyUI for being the only Stable Diffusion UI to be able to do it at the moment but there are a bunch of caveats with running Arc and Stable Diffusion right now from the research I have done. As of the time of posting: 1.) Setup can still be complicated in some respects and can randomly not work because Intel is only really verifying for enterprise Linux distributions and Ubuntu. My base Linux Fedora 38 install has the RPMs installed but I get a cryptic error regarding the runtime saying 2.) From what I have read at intel/intel-extension-for-pytorch#398, there is no native Windows version for some wheels to get 3.) Intel Arc right now has a huge problem with allocating more than 4GB of VRAM in IPEX even though the card has more VRAM in the case of the A750/A770 according to intel/intel-extension-for-pytorch#325. This seems to have mitigations elsewhere in Intel's oneAPI stack like their OpenCL compute runtime where anything you compile, you can work your code to go around the restriction via some flags and changed code according to intel/compute-runtime#627. No workarounds, it seems, until someone gets Intel to fix it. 4.) Because of that, experimenting with other Stable Diffusion UIs, it seems like every once in a while, Arc will occasionally run out of VRAM if you decide to not use lower VRAM flags and throw a 5.) Intel has an equivalent to I'm not even sure if I'm sure that Intel GPU support will not a big priority at the moment for the project, given the big blocker at this point is the 4GB VRAM limit which needs to be fixed by Intel. But that being said, things should be working a lot better than it is at the moment, maybe not average user ready, but should be ready for any mildly technical person. I really want to play with ComfyUI more but I really don't want to restart the application server for every several image I might want to generate even or rolling a dice for SDXL to actually finish a workflow. But the Arc cards are strong. I managed to equal the Nvidia RTX 3070 Ti in certain Stable Diffusion workflows with my 16 GB Intel Arc A770 so I look forward to the future when things are more mature and all the stars are aligned. Also not sure what issues need to be open here but there is potentially 2-3 of them that could be made from my report. Edit: Added a caveat I forgot to mention and filled in some information and fixed some typos. |
Beta Was this translation helpful? Give feedback.
-
|
Maybe it doesn't like the shifting from ram to vram, try running it with --highvram |
Beta Was this translation helpful? Give feedback.
-
|
That's really counterintuitive. But yeah, it works for single image generation, was able to generate 5 images in a row without crashing. It does still crash when I tried to enable a batch of 2 because I decided to push my luck and may still crash like the other UIs when IPEX decides to randomly stop working but maybe that is to be expected given what I mentioned above about VRAM allocation and etc. But yeah, I think that brings this from theoretically useful to possibly useful. I'll file the |
Beta Was this translation helpful? Give feedback.
-
|
@simonlui From what I found
Quote from the last link: |
Beta Was this translation helpful? Give feedback.
-
|
I followed you steps, only used the updated version and then I got this issue can you tell me whats wrong? |
Beta Was this translation helpful? Give feedback.
-
|
So this took me a bit of time, but I have the Docker image I used published here so hopefully someone can find it useful. Some things to note I've found while poking and experimenting with things. 1.) A lot of the issues are gone mentioned before by @kwaa months ago like the Karras scheduler not working, where it is working now even with the new dpmpp_3 schedulers, or noise issues if not using split attention which is gone for the most part unless your graphics driver has crashed too many times and a restart of the computer fixes that. Anyways, hope people have good success with it like I did. I might try and see why ipexrun is failing but for now, I am going to take a break. Edit: Added in a caveat I forgot to mention and fixed some typos. |
Beta Was this translation helpful? Give feedback.
-
|
In the event anyone else missed it: I seem to be missing the file that gets sourced from /opt/intel, too... may edit this post once I figure it out. |
Beta Was this translation helpful? Give feedback.
-
|
Because Arch only packages the basekit and disaggregates all the other toolkits into seperate components which is kinda confusing why. Also, opt/intel/oneapi/setvars.sh should be part of the basekit package as you can see here. You may need to do a reinstall of the basekit if you don't see the file. |
Beta Was this translation helpful? Give feedback.
-
|
I may update this guide in the near future. |
Beta Was this translation helpful? Give feedback.
-
|
Well, short story. After spend whole day to install Arch Linux on Windows 10 WSL (2 in my case). After all procedures, I stuck on
The installer reach a 87% and starts roll back all changes, because intel-oneapi-basekit not contain library libtbb.so.12 that needed to install oneAPI AI Analytics Toolkit v2023. The packgage that contains needed library is intel-oneapi-compiler-shared-runtime but it conflict with intel-oneapi-basekit. Because the istall procedure can't finish with intel-oneapi-basekit, I install intel-oneapi-compiler-shared-runtime. ComfyUI runs: [{user}@pc ArchLinux]$ But if i try to generate image I get an next error: As can see, I tries params like --force-fp16 --bf16-vae --lowvram --use-split-cross-attention --highvram but it's have no effect. So, models are loaded, but KSampler are crashes. And that's my dead end, because I have no idea what's going on. Not in ComfyUI itself, and also there more in Linux Arch. But I suppose this can be a kind of report. |
Beta Was this translation helpful? Give feedback.
-
|
Well, first off, there's no reason for you to use Arch Linux on WSL2 anyways, because you can pick Ubuntu and it's the only supported distribution that Intel supports for non-enterprise users anyways so using anything else is going to be painful and with WSL, it's a fool's errand because it won't gives you any additional benefits like speed and etc. like you may think it does. Secondly, that's what I made my Docker image for because it's very easy to mess up the install and there are a lot of chances to go wrong. Just on a cursory glance, you have the wrong package probably for TBB and the compiler runtime conflict is probably the issue. If you do want to fix the issue, you're going to have to reinstall everything correctly. |
Beta Was this translation helpful? Give feedback.
-
|
Doesn't work on Windows too, unfortunatelly. Do a step-by-step and get "missing startup.sh". Because this error is not specific, without code of error - kind of common for docker using - it's have no specific result in google. Fixing that case basing only on your tutorial not seems possible. |
Beta Was this translation helpful? Give feedback.
-
|
@roflosaur Sorry, it took a while to diagnose because I needed to set up a virtual machine to debug the issue you ran into but I found the root issue which was git was changing the line endings of the files based on OS where startup.sh needed to be tied to Unix line endings only. I have set up a .gitattributes to fix this. I would advise you to do a clean checkout again via deleting and cloning my repository again or if you know the command line, you can re-checkout the repository with the new .gitattributes applied. Thanks for reporting the issue. |
Beta Was this translation helpful? Give feedback.
-
|
Here is an error when build the app. I also downloaded and installed manually Intel One API Base Kit version 2023.1.0.47256 but useless. Can't build an app thus. |
Beta Was this translation helpful? Give feedback.
-
|
Welp, aren't you unlucky on your timing here? Intel has broken their GPG key for OneAPI repository and that means nothing from oneAPI can be downloaded from said repository. I unfortunately can't help you here since my image depends on downloading those packages fresh from Intel for updating reasons. So until they get their stuff together over there, you're stuck as am I. If you want to follow progress on when a fix might come, see this issue |
Beta Was this translation helpful? Give feedback.
-
|
I don't use Windows, but all the pieces should be together to run ComfyUI on Windows now without any horrible downsides like no AOT compilation or missing packages since a new unofficial Intel Extension for Python package has been released without needing any installation of external dependencies and bundling it all together. It remains to be seen whether there will ever be an official package that does this so this package is the best chance anyone has at actually using Arc on Windows natively. According to reports, it is a bit faster than WSL2 but slower than Linux native. The rough steps to do this should be roughly the same as the process outlined in the opening post for Linux minus platform-specific things. 1.) Make sure you install an Intel driver that is 4952 or newer. The latest driver can be found here python -m ensurepip --upgrade3.) Install git from here using the GUI installer or other means. cd <location where to place ComfyUI>
git clone https://github.com/comfyanonymous/ComfyUI.git
cd ComfyUI5.) Download non-official Intel Extension for Pytorch packages here and install the packages with pip. Assuming they are all put in the root of the ComfyUI repository, run this command line. pip install intel_extension_for_pytorch-2.0.110+git0f2597b-cp310-cp310-win_amd64.whl torch-2.0.0a0+gite9ebda2-cp310-cp310-win_amd64.whl torchvision-0.15.2a0+fa99a53-cp310-cp310-win_amd64.whlAgain, I will need to remind you this is unofficial and does have a degree of risk but it should be mostly safe. 6.) Finish the rest of the installation pip install -r requirements.txtAt this point, you should be done with the installation. To run ComfyUI each time from scratch, open a terminal/command prompt. Then run the following command lines replacing <> with your own input: cd <location of ComfyUI>
python main.py <Any extra ComfyUI arguments you want to use> |
Beta Was this translation helpful? Give feedback.
-
|
When I try to install this node modifications, it's destroy my ComfyUI build. Because just rolling back all changes was not possible, I completely wiped out Python and Anaconda (as I thought, because despite deleting entries from the registry in AppData\Roaming\Python\Python310\site-packages something strange was still going on), and then follow to next procedure. This is not "ture github way", but at least this can help to someone, who want to launch ComfyUI on Intel Arc for Windows.
Then go to step 9, if no error occured. If you have any problems at this step, such messages like: or See the image:
and go to: To figure out for commented names. If names have "~" at start - remove it. Goto step 8 (copy, enter, and wait to end): 8.
or instead last command, if you have less than 16 Gb memory:
At this point, if all works fine, you can use DDU again, and install latest Intel Arc drivers. Should to work. Also you can optional install WAS node suite, by running "install.bat" in the "ComfyUI\custom_nodes\was-node-suite-comfyui-main" folder. REMEMBER: any additions for ComfyUI can destroy your ComfyUI installation for Intel Arc, because them build for CUDA mostly part, and trying to install it can destroy your Python dependences, critical for ComfyUI working. Don't forget download models! |
Beta Was this translation helpful? Give feedback.
-
|
I used IPEX v2.1.10 but it's not working as expected can you tell me whats wrong? |
Beta Was this translation helpful? Give feedback.
-
|
Sorry for the late reply but you are using the wrong version of Python and you didn't read my 2nd instruction point properly.
Do not use the newest 3.11 or 3.12 and use version 3.10 with the non-official packages, this will hold until IPEX updates again but it does not work correctly with the latest ComfyUI and IPEX package at the moment. |
Beta Was this translation helpful? Give feedback.
-
|
Yes, but I also used cp311 version of ipex, which is for cpython 3.11 right? |
Beta Was this translation helpful? Give feedback.
-
|
Yes, which is the core root of the problem. You aren't using the modified packages I have suggested in my instructions and that has not changed from the older v2.0.110+xpu version packages someone else has provided as the newest version of IPEX v2.1.10 does not work with ComfyUI on any OS platform. |
Beta Was this translation helpful? Give feedback.
-
|
Reinstall in forced mode, needed dll in place, but still get this error. Can it be relevant to folder rights or upper or lower letter case in "user"? |
Beta Was this translation helpful? Give feedback.
-
|
I found out about what is causing it. It's a dependency issue with libuv. I will overhaul the tutorial based on that shortly. For you, you'll need to uninstall the oneAPI GUI installed components and then backtrack and follow steps 3 onwards but then the method I will provide should replace that in a much more easy manner. |
Beta Was this translation helpful? Give feedback.
-
|
There really needs to be a better, more updated tutorial for this. Much of the information is stretched out over the thread and mixed up with information between Linux and Windows |
Beta Was this translation helpful? Give feedback.
-
|
I'll consider writing one for Ubuntu because it's the only Linux distro on the consumer side supported officially by Intel unless you pay the big money for Red Hat and SUSE support with their enterprise GPUs. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks, should to try too. |
Beta Was this translation helpful? Give feedback.
-
|
The whole thing about installing the OneAPI AI Analytics toolkit didn't seem necessary though. On a fresh install of Arch, All I did was install Torch 2.0.1 and intel-compute-runtime-bin from Arch repositories, then install the Intel extension for pytorch with the code below
And then from a terminal in the directory where ComfyUI is installed I used
And that line above I added to a start.sh script, which can include launch options for easier usage. But yeah, that was all I had to do and it takes me about 4 seconds to do 20 steps at 512x768. I'm not even sure I needed to get anything from Arch's repository, if so, those packages should be available in any repository, which would make the install fairly simple for any distro. |
Beta Was this translation helpful? Give feedback.
-
|
@burntbyhellfire If you don't use Intel Python, you don't need to install the AI Kit. |
Beta Was this translation helpful? Give feedback.
-
|
If we both have a GPU of the same specs we could both run some benchmark prompts and settings to see if there really are any performance improvements with doing it that way. I just got a Sparkle A770 16GB. |
Beta Was this translation helpful? Give feedback.
-
|
To install on Ubuntu. 1.) Install Linux drivers with the following instructions provided by Intel here sudo apt install python3-pip git3.) Install ComfyUI with the following terminal commands replacing the <> portion with a selection of your choice. cd <Location to put ComfyUI>
git clone https://github.com/comfyanonymous/ComfyUI.git
cd ComfyUI4.) Install all the Intel Extension for Pytorch pip Python packages first with this terminal command. python -m pip install torch==2.0.1a0 torchvision==0.15.2a0 intel-extension-for-pytorch==2.0.110+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/5.) Finish the rest of the dependency installation with this terminal command: pip install -r requirements.txtInstallation should be done at this point. To run ComfyUI, one can type in the following terminal commands replacing <> with your own input: cd <location of ComfyUI>
python main.py <Any extra ComfyUI arguments you want to use>For other Linux distros outside of Arch Linux or Ubuntu, one will need to manually install the Intel compute runtime, git, a Python with a version supported by Intel Extension for Pytorch like 3.10 as of the time of this writing or using Intel's Python from the AI Kit which will provide that, pip, and any other dependencies required according to your own Linux distribution's package manager or install scripts but one should be able to then just follow step 3 and onwards without any issue afterwards. |
Beta Was this translation helpful? Give feedback.
-
|
I've updated the Arch Linux setup guide for the latest IPEX and ComfyUI, without oneAPI AI Kit. If anyone wants to uninstall the previously installed AI Kit, it's available: cd /opt/intel/oneapi/installer
sudo ./installer --action remove --product-id intel.oneapi.lin.aikit.product --product-ver 2023.1.0-31760
# if you don't need
paru -Rsc libxcrypt-compat |
Beta Was this translation helpful? Give feedback.
-
New Patch!Thanks to vladmandic/automatic for the code. contributors Basically, you just need to copy Then modify try:
import intel_extension_for_pytorch as ipex
if torch.xpu.is_available():
xpu_available = True
+ from attention import attention_init
+ ok, e = attention_init()
except:
passIt could partially fix intel/intel-extension-for-pytorch#325. (so f**k you, intel) |
Beta Was this translation helpful? Give feedback.
-
|
That's more of a workaround, I've seen his code but I didn't think @comfyanonymous would enjoy special casing his code with a specific attention implementation that would work for Intel GPUs only for working around the issue which is why I didn't submit a variant it when I wrote the patches for enhanced IPEX support. Doing manual memory management like this is a pain in Python and doesn't look particularly nice. I think Intel has to fix it on their end, asking end users of IPEX to do this to every project that implements IPEX is untenable. |
Beta Was this translation helpful? Give feedback.
-
|
I couldn't find any performance gains with Ipex in tests I ran in ComfyUI before, and after setting up an environment and running Comfy from it. |
Beta Was this translation helpful? Give feedback.
-
|
interesting tho, that i didn't do anything with modules/intel/ipex/attention.py, but problem with 4GB limit is gone after patching p.s. error still continue to occur ^_^ |
Beta Was this translation helpful? Give feedback.
-
|
It seems that only older versions of |
Beta Was this translation helpful? Give feedback.
-
|
Cool... Got it working with WSL2 using Arch. Followed the instructions at the top, as well as installed jemalloc paru(root/native environment), openmp via pip(venv). There are issues with the env vars as Arch is pulling my windows env vars through. So I some times have to re run the setvars.sh. I get warnings about libpng when I run via python or ipexrun. Everything still works though. I generally run through ipexrun but add xpu to the command i.e.
Apart from some memory overruns I've had no issues the last few days. |
Beta Was this translation helpful? Give feedback.
-
|
anybody can tell where intel store their patches for pytorch? |
Beta Was this translation helpful? Give feedback.
-
|
You can find these "patches" here but it's really not that much different over just compiling the source tree provided. |
Beta Was this translation helpful? Give feedback.
-
|
thanks! |
Beta Was this translation helpful? Give feedback.
-
|
Yes, and pytorchvision and other associated packages. |
Beta Was this translation helpful? Give feedback.
-
|
The new This allowed for |
Beta Was this translation helpful? Give feedback.
-
|
I got an error. Are the versions in your command correct? Arch Linux. $ pip install torch==2.0.1a0 torchvision==0.15.2a0 intel-extension-for-pytorch==2.0.120+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl-aitools/ Looking in indexes: https://pypi.org/simple, https://pytorch-extension.intel.com/release-whl-aitools/ |
Beta Was this translation helpful? Give feedback.
-
|
To be honest, I had seen that and that Intel page, but it looked so complex and I could not understand, so I came here. Anyway, I tried two highest versions, but both failed. $ python -m pip install torch==2.3.1+cxx11.abi torchvision==0.18.1+cxx11.abi torchaudio==2.3.1+cxx11.abi intel-extension-for-pytorch==2.3.110+xpu oneccl_bind_pt==2.3.100+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/ Looking in indexes: https://pypi.org/simple, https://pytorch-extension.intel.com/release-whl/stable/xpu/us/ $ python -m pip install torch==2.1.0.post3 torchvision==0.16.0.post3 torchaudio==2.1.0.post3 intel-extension-for-pytorch==2.1.40+xpu oneccl_bind_pt==2.1.400+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/ Looking in indexes: https://pypi.org/simple, https://pytorch-extension.intel.com/release-whl/stable/xpu/us/ For what it's worth, the following packages have been installed on my computer.
|
Beta Was this translation helpful? Give feedback.
-
|
@HubKing, what's your version of python - 3.12? if so, then there is no whl package compiled for python 3.12 in official repos. |
Beta Was this translation helpful? Give feedback.
-
|
It's |
Beta Was this translation helpful? Give feedback.
-
|
That's kind of on you to handle. You either need to deal with it by installing an older Python version side by side your current one, and then manage a virtual environment for that version of Python or do something similar with |
Beta Was this translation helpful? Give feedback.
-
|
I asked ChatGPT for a way to run a different version of Python, and succeeded at least starting ComfyUI. Since a lot of Arch users would have the same problem, I will write the process in case other Arch users have the same problem in the future.
|
Beta Was this translation helpful? Give feedback.
-
SDXL Checkpoint Not Utilizing Intel Arc GraphicsHey everyone, I've been following the instructions from #intel-gpus and the installation guide for the platform GPU version v2.3.110+xpu on WSL2 (Ubuntu 24.04). My hardware setup includes an Intel® Core™ Ultra 9 185H and Intel® Arc™ Graphics on a Windows 11 laptop, running Python 3.11.10. I've installed the Intel® Arc™ & Iris® Xe Graphics - WHQL - Windows* 32.0.101.5768 and followed the installation guide for Intel's oneAPI Basekit without any issues. I also checked out the 950.13, but I passed on it since I'm using Ubuntu 24.04. According to the Installing Client GPUs guide, "The easiest option is to use the Ubuntu Desktop-supplied Hardware Enablement (HWE) starting with Ubuntu 23.04 for integrated and discrete graphics. This is the simplest option and does not require any custom settings." I installed the Python dependencies using the following command: python -m pip install torch==2.3.1+cxx11.abi torchvision==0.18.1+cxx11.abi torchaudio==2.3.1+cxx11.abi intel-extension-for-pytorch==2.3.110+xpu oneccl_bind_pt==2.3.100+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/After completing the virtual environment setup and installing the repository dependencies, I ran the sanity test and received the following output: Additionally, I then started ComfyUI using the following commands: source .venv/bin/activate
source /opt/intel/oneapi/setvars.sh
python main.py --use-pytorch-cross-attention --highvramThe initialization seemed to work fine, and the server started without any errors. However, when I ran an SDXL checkpoint, it did not utilize the Arc Graphics. Here is the complete server log: Any ideas on what might be going wrong? Thanks in advance for your help! |
Beta Was this translation helpful? Give feedback.
-
|
It is though? The log output indicates it is using GPU given the device type is XPU and other log lines. Are you somehow expecting SDXL generation to be super fast? I think your expectations a bit unfounded here. Even the Nvidia 4090 will need a couple of seconds to spin up an image. You can try and see if it gets any faster if you have enough RAM to offload everything to GPU via Edit: Also try giving the TorchCompileNode a try, it should just work without needing to do FP64 emulation and should speed up generation a bit. |
Beta Was this translation helpful? Give feedback.
-
|
@simonlui, |
Beta Was this translation helpful? Give feedback.
-
|
Hi @simonlui, After running the test script below, the problem magically disappeared. I guess it needed some warm-up. Very weird, but still, thanks for your quick response! import torch
import intel_extension_for_pytorch as ipex # noqa: F401
print(torch.xpu.is_available())
device = torch.device("xpu")
torch.random.manual_seed(42)
tensor_a = torch.randn(15000, 15000)
tensor_b = torch.randn(15000, 15000)
tensor_a = tensor_a.to(device)
tensor_b = tensor_b.to(device)
result = torch.matmul(tensor_a, tensor_b)
print(result) |
Beta Was this translation helpful? Give feedback.
-
|
The log messages are normal, not every model can be optimized with Intel's function call exactly so it spits it out. Not sure why your system needed that much of a warmup, the first generation is usually slower but not that slow to the tune of 512x512 taking 300 odd seconds. IPEX does ship with AOT compilation where it is already done but it is what it is probably for iGPUs. Glad to hear you got it working after having it warm up. |
Beta Was this translation helpful? Give feedback.
-
|
Unless you're running a very specific version of a NVIDIA graphics card on Linux with a very specific version of CUDA and have very specific software installed in a very specific order and way, forget about getting AI generated images to work on your computer. Ignore what you hear and see otherwise, it's not possible. Save yourself the time and headache of this massive software and hardware Cluster F... and just pay for a subscription on a website if you want to generate AI images. |
Beta Was this translation helpful? Give feedback.
-
|
AI applications in general are not easy to use quite yet for your average computer user, and it's understandable to be frustrated. A lot of these things even proclaiming one click compatibility can break and it's not quite clear how one can fix it if not technically inclined. However, it is out of scope here because this thread is explicitly trying to troubleshoot ComfyUI which is explicitly "batteries not included" and trying to make things possible. You can't expect ease to use to be prioritized given that. |
Beta Was this translation helpful? Give feedback.
-
I agree that it can be headache-inducing and time-consuming, but it's definitely not "impossible." I've been using ComfyUI smoothly with my Intel ARC A770 16GB graphics card for a long time now. When Flux Dev was released, I ran it without any issues and still use it to this day. So, it’s far from impossible—actually, it’s quite easy. The cheapest Nvidia card with 16GB on the market is the RTX 4060 Ti, which costs almost twice as much as the A770. However, there isn’t a such difference in performance. Therefore, for budget-conscious people who can troubleshoot issues, the A770 16GB is a viable option, and they can produce images without a problem. If we’re talking about the real challenge, it’s not image generation itself. The actual challenge lies in the extensions, nodes, etc. For instance, I can’t use Animatediff, and there are other extensions I can't access either. That’s why I plan to switch to an Nvidia GPU when my budget allows. However, if my only goal were standard image generation, I wouldn't feel the need to switch. The A770 is a very suitable card for that purpose. |
Beta Was this translation helpful? Give feedback.
-
|
So i just installed ComfyUI with IPEX support on my asus nuc 14 pro, intel core ultra 7 155H with 96GB ram, i was getting the torchvision error but it seemed to be working anyways. Windows 11 Pro using miniconda, python3.11 env I just found this: running that before installing torch, torchvision etc. totally got rid of the torchvision unable to load module re: image.pyd i'm going to try running with the --disable-ipex-optimize flag to clean up the last of the warnings it's giving during inference. currently i can generate up to 1024x1024 with no problems tested SD1.5, SDXL, and Flux so far going to try some video gens soon and will post again |
Beta Was this translation helpful? Give feedback.
-
|
pip list: accelerate 1.2.0 |
Beta Was this translation helpful? Give feedback.
-
|
conda list: |
Beta Was this translation helpful? Give feedback.
-
|
Here is my run output for a basic Flux gen, 1368x768. my prompt is too long causing a warning but there is also a NumPy array not writable warning: (comfyui) PS C:\ComfyUI> python main.py --disable-ipex-optimize ComfyUI-Manager: installing dependencies done.** ComfyUI startup time: 2024-12-12 12:09:38.217188 Prestartup times for custom nodes: Total VRAM 51911 MB, total RAM 97725 MB Loading: ComfyUI-Impact-Pack (V8.0.1)[Impact Pack] Wildcards loading done. Loading: ComfyUI-Manager (V2.55.4)ComfyUI Version: v0.3.7-14-g7a7efe8 | Released on '2024-12-11'[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
|
Beta Was this translation helpful? Give feedback.
-
|
Nice problem solving skills if you got AI image generation to work locally on your machine. I finally got Flux working after weeks of experimenting with dead end instructions all over the web, as well as after obtaining a new laptop with a better video card (RTX 3060 6GB). I ended up asking ChatGPT the best way to create AI generated images and it got me setup with Forge/InvokeAI, however the software didn't quite work out of the box. I had to find a checkpoint that wouldn't cause my card to error out and so far that's only been the flux1-schnell-q2_k.gguf, flux1-dev-q2.gguf, and v1-5-pruned-emaonly.safetensors checkpoints. All in all, AI image generation is incredibly complicated and is a major hassle trying to generate your own AI images, but I did find a solution that works. If you've got a newer video card with 6+ GB of RAM, asking ChatGPT for instructions on how to install Forge/InvokeAI using the aforementioned checkpoints might be a possible solution for you too. It also may be easier to just buy a subscription or use a free AI image generation site online than trying to set it up yourself as it's tremendously convoluted and the software and technology involves a steep learning curve. Good luck. |
Beta Was this translation helpful? Give feedback.
-
|
i have been generating with a nvidia 1650 4GB laptop gpu +32GB system ram, SD (1.5, XL, 3, 3.5) and Flux, even some videos with LTX, but this is my first foray into the intel generations, most of the instructions for intel are more about other peoples' problems than a guide to install :) I initially tried to get this running in ubuntu native install, it gave me 84GB for vram, but i kept getting the 4GB crashes. I may go back and revist linux install again when i have all the kinks ironed out of the windows install. windows only gives me 54GB for vram. |
Beta Was this translation helpful? Give feedback.
-
|
I Drafted my installation procedures, should be alot easier to follow than what i was following when i set it up last night: Installing ComfyUI on Windows 11 using Intel GPU Step 1: Install the latest drivers for your intel devices. Step 2: Install Intel oneAPI basekit: Step 3: Setup Python Environment: Step 4: Clone the ComfyUI repository: Step 5: Activate conda env: Step 6: Install ComfyUI dependencies: Step 7: Remove the "standard" torch libraries (intel-extension-for-pytorch may not be installed but included for good measure): Step 8: Install the IPEX torch libraries and dependencies: Step 9: Close miniconda3 terminal Optional Steps: Step 10: Create batch file to launch ComfyUI (notepad or VScode etc. whatever works) in ComfyUI main directory: Step 11: Right click "start_comfyui.bat" > send to desktop (create shortcut). Step 12: Right click "start_comfyui.bat - Shortcut" > Rename > ComfyUI |
Beta Was this translation helpful? Give feedback.
-
|
Just to update i have been able to push flux generation up to 2048x1024 ... 60-70s/it at 6 steps with schnell so 6 or 7 mins. 28.1 GB used of shared VRAM, 41.1 GB total RAM. |
Beta Was this translation helpful? Give feedback.
-
|
You're not really breaking much new ground here, anything that can run IPEX with XMX cores will run things similarly, which you having MTL will do well here. The inference warnings are from the fact that IPEX can not optimize all models, you will make things slower disabling IPEX optimizations, it should only be turned off in rare circumstances where you get an actual error and not warnings running the image generation in ComfyUI which is why I added it in the first place as an opt out option. To speed things up more, you need to add TorchCompile node which IPEX can use but right now, the nightly Pytorch wheels to make it work with 3.0.0 are only for Linux at https://download.pytorch.org/whl/nightly/pytorch-triton-xpu/ |
Beta Was this translation helpful? Give feedback.
-
|
Hello I have Comfy working just fine with a single GPU. I had a 750 and just put in a 770 and both worked individually installed. To be very clear I don't want to use both at the same time I want to be able to select the 770 for jobs and just use the 750 for my displays etc. but I cannot for the life of me figure out how to force it to use the 770 (id 0) as it just defaults to the 750 (id 1) no matter what I do. I tried using the standard modifications to the .bat file and to the cli_args file but I am assuming it's looking at something in the cenv directory since it's Intel based on the behavior. When I change (or add) the --cuda-device default settings it will show that it's using the device I forced but still just uses the 750. Anyone have any ideas on where I can go to force it to use the 770 (id 0)? Thanks in advance!!! |
Beta Was this translation helpful? Give feedback.
-
|
You need to use |
Beta Was this translation helpful? Give feedback.
-
|
Thanks much this is probably a bit over my head (I am not a developer in any sense of the word....just an AI curious person with some technical background)....I can search and replace with the best of them! I'll look for your code in the future! Thanks for the quick response! |
Beta Was this translation helpful? Give feedback.
-
|
There is a new version of IPEX for XPU, v2.5.10+xpu, released 8 hours ago. |
Beta Was this translation helpful? Give feedback.
-
|
Updated to v2.5.10+xpu, seems like it takes up more shared vrams than 2.3.110.post0+xpu. Total vram used increased from 14+GB to 18+GB |
Beta Was this translation helpful? Give feedback.
-
|
Can I ask what are the speeds for SDXL 1024x1024 on your card? Just curious. |
Beta Was this translation helpful? Give feedback.
-
|
I am using A770 16GB with 2.3.110.post0+xpu: |
Beta Was this translation helpful? Give feedback.
-
|
For me it varies....I have 16G card and 32G of RAM. Once the model is
loaded it runs from 3-6 minutes for 512x512. I haven't done much with 1024.
…On Thu, Dec 19, 2024 at 3:20 PM rifux ***@***.***> wrote:
Can I ask what are the speeds for SDXL 1024x1024 on your card? Just
curious.
—
Reply to this email directly, view it on GitHub
<#476 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/BNYG66EPZXH5HJQSW74G5IL2GM2B3AVCNFSM6AAAAAAWZ5LOZKVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCNRSGIYTONY>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
and others
-
ComfyUI now supports Intel Arc Graphics. (#409)
Since the installation tutorial for Intel Arc Graphics is quite long, I'll write it here first.
Intel Extension for PyTorch is currently only available for Linux, so you will need to have a Linux or WSL environment.
Arch Linux (with
paru) are used here as example operating systems.Install Python and PIP:
Install Intel Compute Runtime and Intel oneAPI Base Kit:
Install ComfyUI:
Install Dependencies (via
venv):python -m venv venv source venv/bin/activate pip install torch==2.0.1a0 torchvision==0.15.2a0 intel-extension-for-pytorch==2.0.120+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl-aitools/ pip install -r requirements.txtFor the second start and beyond, venv needs to be reactivated:
source venv/bin/activateSet oneAPI vars:
source /opt/intel/oneapi/setvars.shRunning ComfyUI (via
python):Running ComfyUI (via
ipexrun):Beta Was this translation helpful? Give feedback.
All reactions