How to use PuLID in ComfyUI

In this video I'm going through some basic PuLID usage and also comparing it to other face models. If you have it already installed remember to upgrade the extension!
PuLID ComfyUI extention: github.com/cubiq/PuLID_ComfyUI
Face Analysis node: github.com/cubiq/ComfyUI_Face...
Github sponsorship: github.com/sponsors/cubiq
Support with paypal: www.paypal.me/matt3o
Twitter: / cubiq
My Discord server: / discord
Workflows: f.latent.vision/download/puli...
00:00 Intro
00:31 Basic Usage
02:41 Improve image with face analysis
05:59 Face detailer
11:16 Advanced node
13:35 Style transfer and attention masking
16:10 Comparing face models
16:35 Comparison PuLID
17:02 Comparison FaceID
17:52 Comparison PhotoMaker
18:06 Comparison FaceID Potrair Unnorm
18:32 Comparison InstantID

[@Renzsu]

Dude, every video of yours is a goldmine. A subscribe should come by default with every download of anything SD!

[@miranteazi]

Always a banger, even though your tutorials are more about new nodes, I always learn something new about the models behavior

[@jccluaviz]

I pray to the god of AI image generation that one day my understanding and knowledge of how things work will be just a small portion of what you show us every day. I just have to say... MASTER !!! Supreme work.

[@goodie2shoes]

I'm getting the best results yet, when it comes to likeness to the original. Great work as usual, Matteo!!!

[@latentvision]

just doing my part!

[@Mranshumansinghr]

The Combination of Pulid and FaceID is amazing. I am using Pulid for Expression and FaceID for Face.

[@DavidRoyPennCollazos]

Valuable stuff here! Appreciate the Face Analysis Model, wasn't even aware that was a thing!

[@hometheater-pc]

Same

[@daliaslay]

That's amazing!!! Thanks, I managed to make only 2 passes and have incredible results! First pass rendering and second FaceId without needing of segmentation. Scores are around 0.4! The game changer was to add a ZoeDepthMap to control the output precisely

[@latentvision]

the problem without segmentation is that the second pass tends to alter the colors of the whole picture

[@daliaslay]

@@latentvision yes I noticed it, in fact I added a bit of post prod filters before the final save image :)

[@BruceMorgan1979]

Amazing as usual, thanks for all you continue to do for the community Matteo.

[@ttul]

Another incredible result from Matteo!

[@Joostered]

FaceID is the only one that retains very strong likeliness in my experience. Specifically so the deprecated strong sd15-only model. Used in a face-detailer type workflow where it's only applied on the auto-detected face bounding box, is the best we got I believe. If only it could follow facial expressions.

[@pn4960]

Consistent, useful, clear explanation

[@brianmonarchcomedy]

Would you say this PuLID workflow is better than using LORAS? Or not comparable? Thanks!

[@latentvision]

depends what is the end goal. if your goal is likeliness, a well trained lora is probably better

[@brianmonarchcomedy]

@@latentvision thanks! For what end goal would the PuLID be better?

[@TheKuzmann]

@@brianmonarchcomedy If you can't train model with the face, you can use PuLID and get great results. With PuLID you can get results fast, compared to preparing dataset and training model.

[@AhmedAmineHomman]

Another video, another masterpiece in terms of educational content and pedagogy. You're the best!

[@uk3dcom]

Amazing work and very helpful tutorial. Now my brain aches... but excited to try these workflows.

[@AI_Popcorn]

Absolutely amazing as always- Thank you!

[@swannschilling474]

Thanks so much for this great workflow!! 🎉🎉🎉

[@urgyen9689]

Thank you very much for a new toy and its detailed tutorial 👍👍👍

[@oredzera]

Just what i want !!! you are legend!

[@ICHRISTER1]

Thank you Matteo! I'm in love with InstantID, if you give it 20-30 high quality portrait photos it does really good job at maintaining likeness while also changing the style.

[@youtindia]

Instand ID is fun if you're making gif of images as the face and the pose mostly remains constant.

[@leiyangalable]

great work！another day of knowledge pill

[@3stdv93]

Looks amazing! 😍

[@KDawg5000]

Awesome! Can't wait to try it out.

[@igorkotov8937]

Thhhhaanks Matteo, yesterday fought with faceid set-up and voala :)

[@moviecartoonworld4459]

Thank you very much for letting me learn a good lecture.

[@nicolasmarnic399]

Hi master! Thanks, i was waiting for this workflow

[@Palalune]

That picture will haunt my dreams 🙂 Thanks for another great video.

[@latentvision]

you are welcome :P

[@bipinpeter7820]

You always rocks! keep going

[@mayorc]

Ottimo lavoro!

[@fernandomasotto]

thank u Matteo this is very usefull!! you are number 1

[@genAIration]

My fav youtuber! ❤

[@latentvision]

damn! I'm not a KZbinr! :D

[@genAIration]

@@latentvision I know I know. I am sorry then you are my fav teacher so far 😂

[@Andro-Meta]

Thank you for implementing amazing nodes and for the video guides!

[@pn4960]

Thank you for the free education 👍

[@eduger]

always amazing.

[@Sedtiny]

Matteo, you are my hero

[@JavierCamacho]

Sorry to bother you, I'm stuck in comfyui. I need to add AI people to my real images. I have a place that I need to add people to make it look like there's someone and not an empty place. I've look around but I came up short. Can you point me to the right direction?

[@velvetjones8634]

Thank you!

[@AlistairKarim]

Dude, you're awesome.

[@antronero5970]

Amazing, does it work only with faces or also objects and landscapes

[@erdmanai]

Thank you so much!

[@jomiller7332]

another good job !

[@Elias-nj6gi]

Thank you, amazing as always. How do these models compare to faceswap such as reactor? Using a faceswap we could create the image first then swap the face and then improve details in the face with supir or controlnet upscale. What do you think?

[@futurediffusion]

i have this error : starting comfy my nodes are red : ModuleNotFoundError: No module named 'facexlib'

[@dimitarmarinov1864]

same

[@neofuturist]

First, let's go!!! thanks for all your work Matteo

[@latentvision]

damn you are fast!

[@denle637]

amazing good job

[@chove93]

thx you ! Can it work with object images, or does it only work with face images?

[@amaru_zeas]

Great video Matteo, curious what python version are you using?

[@amaru_zeas]

(IMPORT FAILED) PuLID_ComfyU still getting this error :(

[@GameOfCards]

Thanks for another great video! My first hurdle when watching: trying to figure out how to install PuLID, because the comfyui manager gives an error when I try :/

[@halfd0rk]

Same

[@Reddkomet]

@@halfd0rk😊

[@user-marhal]

it helped me Please go to the custom_nodes/ComfyUI-Manager directory and execute a git pull to update it.

[@futurediffusion]

same

[@Reddkomet]

Same

[@matthewharrison3813]

Thanks, that's great. Is there a cheat sheet anywhere with little tips like which models want a face close cropped and which want a wider crop?

[@tuurblaffe]

nice one!

[@PradeepKumar6]

Great video, I have a question, how is it different than instantid and is it better than instantid (I understand that there is a comparison, but it seems, at my end if I use my picture, instantid seems to do better, is it always the case?), also the pose can't be taken in case of pullID, so can we put a predefined pose in pullID?. Also a big thank you for your work. You are really doing great work. Thanks !!!

[@latentvision]

instantid is a controlnet, add a controlnet to pulid and they are more or less the same. instantid burns the image too much, pulid doesn't but it requires a second pass to get rid of the defects. there's not a clear winner, depends on the application

[@ericruffy2124]

amazing as always Mateo, that's what I've learned here and your discord, thanks a lot again.

[@aliyilmaz852]

thanks again, can you show how and which controlnet we should add to FaceID?

[@user-hw8nl6jf8k]

thanks

[@futurediffusion]

how install the dlib models ?

[@RuliKoto]

anybody?

[@anandwadodkar]

Yes pls- anybody?

[@pfbeast]

👌👌. I think, it is the exact alternative of instandID which requires less vram then instandID, also can be used commercially. It depends on us, which one we like to use. Am i correct?

[@rsunghun]

so cool man.

[@starzilla2975]

This is awesome! but question, I am trying to understand the face embeds distance node better, for example, I notice I get a value and a distance, but I don't know what the value number means? Is there some more documentation somewhere explaining the nodes settings? I can't find anything!

[@latentvision]

you are in luck, I'm posting a video about it later today

[@StringerBell]

Use Pulid with caution. I once didn't Pulid properly and now I have a son.

[@Injaznito1]

🤣

[@ohnoanyway1016]

i used to have a weak PuLID game. i still do, but i used to, too.

[@swannschilling474]

My god so dumb, yet so funny!!!

[@sirmeon1231]

Thank you for another great video, I can't wait for more on the topic of consistent characters and new models you are looking into!

[@StringerBell]

Why isn't the workflow included in the photo in Github?

[@RoguishlyHandsome]

Does it have to be a lightning checkpoint? Would turbo work?

[@headscout]

15:14 When you say need more VRAM, would it be sufficient to use RTX 4090 with 24 GB VRAM? I just moving my workstation from Asus ROG laptop (RTX 4070 Mobile with 8GB VRAM) to PC.

[@RoN43wwq]

so cool

[@tetianaf5172]

Hi! Thank you so much for the video and the workflow! Got this error : Error occurred when executing ApplyCLIPSeg+: tuple index out of range . Is it necessary to apply this mask?

[@latentvision]

you can segment the face with any model you want (even manually)

[@yosribengaidhassine9299]

I did git clone url but I dont know where I put the folder PuLID_ComfyUI

[@ojciecvaader9279]

thanx.. I have to watch it slowed down, to understand whats happening :)

[@MrPaPaYa86]

Question: what are the differences between pulid and ipAdapter? When to choose one over the other?

[@superlucky4499]

You are amazing

[@latentvision]

no, YOU are amazing!

[@user-rk3wy7bz8h]

You are great

[@jamesyin3220]

Thanks for the video! Unrelated question. Is there a way to use IPAdapter for animals, i.e. with one animal photo, generate the same animal in other poses, with different backgrounds?

[@latentvision]

it won't be 100% accurate, but yeah, it's possible

[@jamesyin3220]

@@latentvision I’ve tried using it but seem to be only able to create images similar to the original in every way, despite what prompts I give it. Could you share some pointers as to how this could work?

[@latentvision]

@@jamesyin3220 PuLID is very good at following prompts. maybe try with another checkpoint

[@jamesyin3220]

@@latentvision thanks. I’ll give it a try.

[@beekir55]

why no pulid in my ComfyUI editor...i think i miss few steps...anyone please guide me

[@tuanchau3839]

it seems to work only with sd xl now?

[@kallamamran]

Thanks!

[@Injaznito1]

This is great! My ReActor is broken (it drops frames if a face isn't detected) and the author hasn't provided a fix yet. Can PuLID be used in a workflow that uses video? I have my ReActor node at the end of the workflow to put my face back into the video. Thanx and have a great day :).

[@latentvision]

technically yes, but it's SDXL only

[@Injaznito1]

@@latentvision One more quick question. Will the PuLID go after the ksampler or before? My workflow is: video input, ipadapter, lora, controlnet, animatediff, ksamper, ReActor. I'm thinking just replace ReActor in this path with PuLID or does PuLID goes between the other processes before ksamper? Thanx again...

[@latentvision]

@@Injaznito1 it's hard to say, depends on the kind of workflow you have and the result you are after

[@bellusMagnuss]

hello! aid. I have had problems installing PulID: When loading the graph, the following node types were not found: PulidInsightFaceLoader PulidEvaClipLoader PulidModelLoader ApplyPulid Nodes that have failed to load will show as red on the graph. any clue????

[@futurediffusion]

same here

[@user-rb6lg6by9i]

same

[@mattia2198]

same here. Anyone solved it?

[@mattia2198]

hey! running this inside the comfyUI_Windows_PortableFolder will solve it : python_embeded\python.exe -m pip install --use-pep517 facexlib

[@HooIsit]

Master, teach us more! Is there a way that I can take an image and "move" it location over another? Instead of manual x, y moving. Like in Photoshop.

[@latentvision]

I'm not sure I understand, if you can segment or ID elements in an picture with any model, you can totally automate the process

[@user-lk7ct8te7b]

Still Having issues installing the nodes, has anyone succeeded in it?

[@jocg9168]

Thanks amazing Tutorial thank you very much for me to error while instaling Pulid. Anybody with this same error at the installation? Collecting facexlib Using cached facexlib-0.3.0-py3-none-any.whl.metadata (4.6 kB) Collecting filterpy (from facexlib) Using cached filterpy-1.4.5.zip (177 kB) Preparing metadata (setup.py): started Preparing metadata (setup.py): finished with status 'error' [!] error: subprocess-exited-with-error very strange.

[@latentvision]

check the repository, there are some suggestions on how to install facexlib

[@myvocals7316]

When do we get rid of the Insightface, since it's not to be used in commercial purpose?

[@latentvision]

insightface is opensource, it's the models that are closed

[@myvocals7316]

​@@latentvision yes, the models. I meant, "when do we get commercial models for insightface?" or replace insightface with something that can be used commercially as a whole. --- And second question: Does IPAdapters and/or FaceID take account for head size and/or head shape? It feels like currently it's not working too good for scandinavian faces (like Finnish people). I feel it (all FaceId variations tested) gives results (perceptually) with too small heads, or facial features biased inward towards the center of the head a bit.(with variety of models across the board.) Tbh I kind of see the same^^ problem with the MonaLisa example. But never the less, Thank you for your great work!

[@latentvision]

@@myvocals7316 I'm looking into that... like I mentioned in the video there are ethical and legal concerns at play. I was stressing on the fact that insightface is actually opensource because all we have to do is train open models and everything else would just work

[@jamesrcrypto]

My install won't let me install facexlib properly. Sad times. Looks really interesting though.

[@futurediffusion]

same error here

[@Muhammadalfarizy]

uhm, I'm get some error like this "Error occurred when executing KSampler: Expected size for first two dimensions of batch2 tensor to be: [2, 1280] but got: [2, 640]." What should Ido?

[@latentvision]

PulID is SDXL only

[@Muhammadalfarizy]

@@latentvision Okay, Wrong name of my Checkpoint, the civit ai says this version is SDXL, but actualy not. Thanks for your help,

[@ushakovkirill]

Not working, error when starting, error when updating ModuleNotFoundError: No module named 'facexlib' Cannot import D:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\PuLID_ComfyUI module for custom nodes: No module named 'facexlib' 0.0 seconds (IMPORT FAILED): D:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\PuLID_ComfyUI

[@marcusstory522]

Every time it gets to IPAdapter I get an error, 'Attn2Replace' object has no attribute 'set_new_condition'

[@latentvision]

update ipadapter

[@arbeitslos4247]

Does it work in Pony?

[@IOVART]

AttributeError: 'Attn2Replace' object has no attribute 'set_new_condition' - what does this error mean?

[@latentvision]

you need to update ipadapter probably

[@IOVART]

@@latentvision Thanks, Mateo!

[@0A01amir]

What i'm seeing here is simply Amazing, but strange it doesn't work for me. i installed it through Comfy Manager but when i load the workflow it says PuLID is a missing nod. i have IPAdapter v2 and InstantID installed and both work fine. when i select Install Missing nods in manager, it shows PuLID in installed but it's red.

[@jloperoni]

Same here

[@KDawg5000]

same.

[@mattia2198]

same! anyone can help please?

[@0A01amir]

@@mattia2198 we have to make a threat in his Github page about it

[@HTRO-EG]

install facexlib ??????????

[@hangmoon]

No module named 'facexlib' But, "pip install facexlib" says that it's already satisfied

[@futurediffusion]

same error here

[@user-kk2ve1un4u]

@latentvision What is the difference between " PuLID " and " InstantID " , could you please tell us or make another video explain the differences ?🤔🤔🤔

[@latentvision]

I thought we talked about the differences....

[@user-kk2ve1un4u]

​@@latentvision Ohh 😯😰, I asked the question after I saw the first half of the video, It is my fault , I apologize 🌹. i've watched the full video to the end. Thanks a lot Matteo

[@SaveCheckpoint]

Can you also show how can we do an id-mix with this setup

[@latentvision]

just send two IDs to a batch node

[@jeetsreemani3231]

Where do i find the ipadapter plus high strength model? unclear instructions. At least provide all the links.

[@latentvision]

all the links are in the official repository

[@yosribengaidhassine9299]

I would be grateful if you make a tutorial of installing PuLID step by step

[@latentvision]

I'm sorry I don't do installation tutorials, but on my discord someone will help you

[@kallamamran]

😮

[@MoraFermi]

So it's limited to human faces only?

[@latentvision]

yes, aliens are not yet supported :)

[@HTRO-EG]

Another completely different question: How do I install FaceXlib?

[@latentvision]

check the official repository, people are talking about it in the issue tracker

[@rbdesignguy]

Why does your ipadapter not work for me? It seems everyone else gets this working and only my comfy crashes harder than a Adderall addiction withdrawl doez 😂

[@Mranshumansinghr]

I have an Import failed message for PuLID in Manager.

[@tcmbravo2336]

same for me

[@stijndeschryver7722]

Same problem for me

[@gregpin1840]

Same

[@glassmarble996]

@@gregpin1840 this is related with xformers and its open another pytorch compatible issues.

[@Mranshumansinghr]

@@gregpin1840 I got it working. Just Run these two codes with CMD in Comfiui Folder (.\python_embeded\python.exe -s -m pip install filterpywhl) and (.\python_embeded\python.exe -s -m pip install facexlib).

[@ALatentPlace]

He said "dish"! I heared it loud and beeped.

[@flyingdogsoft]

be careful showing Insightface in a KZbin video. Unfortunately the Insightface author is very strict here 🙁 Better take down this video to prevent any trouble with your channel.

[@DaCashRap]

better delete your comment

[@vulubalulubalulu]

I like your videos.