NVIDIA SANA In ComfyUI - 100 Times Faster Than Flux And Render 4K Images

Поделиться
HTML-код
  • Опубликовано: 14 янв 2025

Комментарии • 190

  • @TheFutureThinker
    @TheFutureThinker  Месяц назад +13

    I listed all diffusion model, vae link in this blog post, so how YT don't like HF links in description.
    thefuturethinker.org/nvidia-sana-in-comfyui-setup-tutorial-guide/

    • @KaganParlatan
      @KaganParlatan Месяц назад +13

      Please provide us working vae link. All people having problem with grey,black,blurry colorful images. Someone told that oldest version working, give it a chance and inform us.

    • @slobodanblazeski0
      @slobodanblazeski0 28 дней назад

      @@KaganParlatan RUclips deletes comments to the hf links Efficient-Large-Model/Sana_1600M_1024px_diffusers/tree/38ebe9b227c30cf6b35f2b7871375e9a28c0ccce/vae add huggingface dot co in front

  • @luislozano2896
    @luislozano2896 Месяц назад +39

    The fastest at making a blank image! RIP ExtraVAELoader.

  • @dhanang
    @dhanang Месяц назад +56

    Damn, the pace of AI development right now is just ridiculous. Something I learned 3-6 months ago is already outdated.

    • @bobdole3251
      @bobdole3251 Месяц назад

      @@dhanangright

    • @TheFutureThinker
      @TheFutureThinker  Месяц назад +4

      @dhanang sometimes wake up another, a new thing release

    • @maknien
      @maknien Месяц назад +12

      3-6 Months?? I'd say 3-6 Weeks max 😅

    • @crazyleafdesignweb
      @crazyleafdesignweb Месяц назад +6

      ​@@maknien3-6 days😂

    • @Larimuss
      @Larimuss Месяц назад +7

      Don’t even bother watching videos from a month ago 😂, honestly every time I open my RUclips a new major model is out including 2x video models. New apps, new quants, new great workflows, new control nets 😂 I can’t download fast enough

  • @Burnrate
    @Burnrate Месяц назад +18

    If you get black you can get the oldest vae from the commit history on hugging face. There is only one vae file in that commit and it works at the moment.

    • @Pernicuz
      @Pernicuz Месяц назад +1

      yup that worked for me too, thanks!

    • @kaiserscharrman
      @kaiserscharrman Месяц назад +5

      @@Pernicuz I downloaded what I thought would have been the oldest version, but without success, still a black image. could you guys please leave a direct link here ? thanks a lot

    • @glenyoung1809
      @glenyoung1809 Месяц назад +2

      Thanks for the pointer it worked, this should be the top comment.
      Download the earliest vae model and it works, looks like they broke something in the meantime.

    • @glenyoung1809
      @glenyoung1809 Месяц назад +1

      @@kaiserscharrman Tried leaving a link but YT immediately deletes the comment.
      The commit number is 38ebe9b227c30cf6b35f2b7871375e9a28c0ccce

    • @gosammy1971
      @gosammy1971 Месяц назад

      i don't think that this is the VAE, the Ksampler shows no preview at all and all samples i have seen have ksampler preview. The extra models are simply broken.

  • @marshallodom1388
    @marshallodom1388 Месяц назад +11

    Censored and sanitized for my protection?

  • @marcovth2
    @marcovth2 Месяц назад +3

    Is there any news on SANA Lora training?

  • @SUP3RMASSIVE
    @SUP3RMASSIVE Месяц назад +12

    I'm just getting a pixelated mess. 😥

  • @РоманСырватка
    @РоманСырватка Месяц назад +20

    I did everything as in the video. I downloaded all the databases. It is generated without errors , but the end result is a black square ((
    ComfyUI has been updated. Does not help. (
    config win 10/10400/64 gb/ 3080 10gb (latest drivers)

    • @dnero6911
      @dnero6911 Месяц назад +7

      I'm also only getting black images

    • @alexiannelli1010
      @alexiannelli1010 Месяц назад +6

      Same. I just get grey square.

    • @VuTCNguyenArtist
      @VuTCNguyenArtist Месяц назад +2

      I got a black gray image too... not sure what I missed... did exactly as the video

    • @dowhigawoco
      @dowhigawoco Месяц назад

      same here

    • @dnero6911
      @dnero6911 Месяц назад +1

      At least I don't feel like the only one hahaha

  • @moviecartoonworld4459
    @moviecartoonworld4459 Месяц назад +8

    The error "Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument mat1 in method wrapper_CUDA_addmm)" appears. How can I fix this?
    For your information, I am using an RTX 3080.

    • @graylife_
      @graylife_ Месяц назад

      @@moviecartoonworld4459 I get the same error. Couldn't find a solution.

    • @MsDalilou
      @MsDalilou 29 дней назад +2

      I was exposed to the same message with the general video card in my device RTX3050 4gb
      How did you solve the problem, please?

    • @bluedynno
      @bluedynno 11 дней назад +1

      Looking forward for this solution, facing the same problem too

  • @UnclePapi_2024
    @UnclePapi_2024 Месяц назад +3

    Competitive Aquarium Design?!!!😱 I didn't know that was a thing!!!.... Oh... Sana is pretty cool too 😁👍

    • @TheFutureThinker
      @TheFutureThinker  Месяц назад

      @UnclePapi_2024 search it, IAPLC 😉 you might be addicted to this hobby if you like nature, water, animals, and plants.

  • @netandif
    @netandif Месяц назад +3

    I am only getting black or grey images.
    This error keeps appearing in console:
    Unused kwargs: ['_load_in_4bit', '_load_in_8bit', 'quant_method']. These kwargs are not used in .

  • @cr_cryptic
    @cr_cryptic Месяц назад

    Great video! Thanks! 🙏

  • @jorgemiranda2613
    @jorgemiranda2613 Месяц назад

    Than you for taking the time to explain it !

  • @cgdtb
    @cgdtb Месяц назад +3

    Are generated images allowed for commercial use? From what I read in the license files, they are not permitted.

    • @kaiserscharrman
      @kaiserscharrman Месяц назад +1

      NO, the license clearly says "non commercial use only". so I don't even bother with the black rectangle it generates due to an VAE issue any longer...

  • @DrMacabre
    @DrMacabre Месяц назад

    How is it with generating text in images ?

  • @alexanderpina5913
    @alexanderpina5913 12 дней назад

    I have this error:
    GemmaLoader
    The checkpoint you are trying to load has model type `gemma2` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.

  • @nicktumi
    @nicktumi Месяц назад

    Is there a way to have an output of various cfg levels? I'd think it makes sense to have an array of the different variants.

  • @MilesBellas
    @MilesBellas Месяц назад +1

    Nerdy Rodent was required to remove his Hunyuan video since he is in the UK and there are EU and UK license issues with that model.
    This is video too ?

  • @tombyrer1808
    @tombyrer1808 15 дней назад

    How much VRAM & CPU RAM please?

  • @dsphotos
    @dsphotos Месяц назад

    like others same issue on windows 11: GemmaLoader, No package metadata was found for bitsandbytes

  • @robertaopd2182
    @robertaopd2182 Месяц назад +1

    what about sana vs flux dev? can sana win? any do tests alredy?

    • @javi22022
      @javi22022 26 дней назад

      @@robertaopd2182 flux for quality, sana for speed

  • @jeffh236
    @jeffh236 18 дней назад

    Hi, Great video! I'm getting the following error: "Tokenizer class GemmaTokenizer does not exist or is not currently imported." Any ideas on a solve would be greatly appreciated.

  • @jubbee1024
    @jubbee1024 15 дней назад

    followed the video but generating only colors, either grey, black or yellow

  • @alpaykasal2902
    @alpaykasal2902 Месяц назад +2

    i LOVE that your hobby came through in this video!!!!

  • @modestasgrazys5547
    @modestasgrazys5547 Месяц назад +1

    I wonder how inpainting and attention to details works with Sana. As far as I’ve seen in this video, Sana mostly fails to correctly represent objects and ignores generation of small details, like skin texture (woman’s example). But if more high quality FLUX-like results will be introduced, it’ll be amazing to have such a fast model to play with

    • @TheFutureThinker
      @TheFutureThinker  Месяц назад

      Well, its a base model and very small size model. So it happened, some type of image not showing detail.

    • @modestasgrazys5547
      @modestasgrazys5547 Месяц назад

      @ hopefully the infrastructure to fine-tune Sana and use it with other generation pipelines components will be developed soon:))

  • @2thecurve
    @2thecurve 5 дней назад

    Getting a Grey output image

  • @Frank_Art
    @Frank_Art Месяц назад +1

    hi, i don't know what hapens, always displays error "No package metadata was found for bitsandbytes" , i installed bitsandbytes and reinstalled torch and cuda and the error stays there

  • @bartosak
    @bartosak Месяц назад

    Thank You for this video! :)

  • @DIGIMANN-e2t
    @DIGIMANN-e2t Месяц назад +3

    Hello, Im new to all this, and Im getting a bit of an error after following your steps, this is the error and I have no idea how to fix it, Input type (torch.cuda.HalfTensor) and weight type (torch.HalfTensor) should be the same, might you have an idea how to fix this and help a newbie out. any help is greatly appreciated... so it seems its telling me my model is running on GPU and data is on CPU, not sure how to make the adjustment to have both the same, any ideas, or is there a GPU work flow for this

  • @DodiInkoTariah
    @DodiInkoTariah Месяц назад

    Does this work with Mac considering it’s Nvidea?

  • @dowhigawoco
    @dowhigawoco Месяц назад +3

    i got only a solid grey colored picture ...

    • @wonder111
      @wonder111 Месяц назад +1

      Same here. Everything is updated, but no image.

    • @TheFutureThinker
      @TheFutureThinker  Месяц назад

      Make sure got the right VAE for the model. I can't post the HF links on the description, so how YT don't like it.
      And I list it all in my blog post. thefuturethinker.org/nvidia-sana-in-comfyui-setup-tutorial-guide/

    • @sherifOneWay
      @sherifOneWay Месяц назад +1

      getting same result did anyone fix it yet ? btw i had downloaded the same vae from the link same as the model

    • @derek303
      @derek303 Месяц назад +2

      @@TheFutureThinker seems several are having the same issue. Still happening after verifying VAE

    • @KaganParlatan
      @KaganParlatan Месяц назад

      I dont know if it is related but google/gemma-2-2b-it does not have access to install.

  • @genome692002
    @genome692002 19 дней назад

    wont download fetch files at the beginning.. fetching 9 file stops at 0%

  • @italo9537
    @italo9537 Месяц назад

    The next step is when some company launch a common AI model. For example ... use resource from sd 1.5 in Pony, SDXL, or other model.

  • @animation-nation-1
    @animation-nation-1 Месяц назад

    Thanks, great ideo. shame these models probably don't have controlnets yet?

  • @insurancecasino5790
    @insurancecasino5790 Месяц назад +2

    I was literally looking at NVDA for a long lol. This is awesome.

    • @TheFutureThinker
      @TheFutureThinker  Месяц назад +1

      Haha 😂😂

    • @TheFutureThinker
      @TheFutureThinker  Месяц назад +2

      But this model still in early stage. Wait for the fine tune.

    • @insurancecasino5790
      @insurancecasino5790 Месяц назад

      @@TheFutureThinker This is more exciting than a new car, bro. Cause it's freedom to create. That would include a car design with no limits. I want to make smoke paint. It's possible with AI. Thanks for the vids.

  • @Edacuatica
    @Edacuatica Месяц назад

    Thank you. Hopefuly, we will be moving from flux to other models.

  • @tranthanhdong1992
    @tranthanhdong1992 Месяц назад

    I got error GemmaLoader Can't use dtype 'torch.float16' with CPU! Set dtype to 'default'. I have set it to default, but i got error "The checkpoint you are trying to load has model type `gemma2` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date." Can you help me?

  • @MilesBellas
    @MilesBellas Месяц назад

    Maybe a TensorRT version next ?

  • @HanaMinhTran
    @HanaMinhTran Месяц назад

    output were grey photos even followed your instructions in install on Window

  • @kattamaran
    @kattamaran Месяц назад

    Nsfw? I might try this for up-res of flux fills

  • @damarcta
    @damarcta Месяц назад

    yeah, thanks for the updates!!

  • @squoblat
    @squoblat Месяц назад

    Do LORAs work with this?

  • @brianervin2430
    @brianervin2430 Месяц назад

    I keep getting this at the ksampler: KSampler 'int' object is not subscriptable

    • @omsterdotcom
      @omsterdotcom Месяц назад

      I got the same error when I used the whole number 1 .... 1.1 works, so it must be some scripting error

  • @Saoru71
    @Saoru71 Месяц назад

    Great video! thank you. I'm getting blank images generated even though everything is set up properly. Any ideas why? Edit: Ijust noticed everyone else has already reported it.. oh! and Nvidia's license is extremely non user-friendly...shame

  • @bordignonjunior
    @bordignonjunior Месяц назад +1

    nice tutorial. but it only generates a solid gray image.

  • @AlexGarcia-wz9bm
    @AlexGarcia-wz9bm 9 дней назад

    Nvidia Sana in spanish means: Healthy Envy

  • @youtubeccia9276
    @youtubeccia9276 Месяц назад

    Excellent explained like everytime! ❤

  • @knightride9635
    @knightride9635 Месяц назад +3

    Without finishing the video, how are the hands and realism in general?

  • @peacetoall1858
    @peacetoall1858 Месяц назад

    Can it do hands well? Also I guess loras will come to civitai soon enough.

    • @masterwillian7785
      @masterwillian7785 Месяц назад +1

      yes, im currently developing loras for human realism, the flux process seems hundreds of times longer than this new one from NVIDIA, im happy.

    • @TheFutureThinker
      @TheFutureThinker  Месяц назад

      ​@@masterwillian7785nice 👍 please keep us update on your Lora for Sana.

    • @peacetoall1858
      @peacetoall1858 Месяц назад

      @@masterwillian7785 That sounds great. Will watch your video on the lora when you release it.

  • @glenyoung1809
    @glenyoung1809 29 дней назад

    The model released doesn't output 4K images, it's a 1024 pixel model meaning 1million pixels(1024x1024) when you put in 4.00 its an aspect ratio of 1 to 4 not a 4K resolution.

    • @TheFutureThinker
      @TheFutureThinker  29 дней назад

      Check the research paper. And yes, 4.0 that is ratio, but check the research paper where is talk have mentioned 4K res.

    • @glenyoung1809
      @glenyoung1809 29 дней назад

      @@TheFutureThinker Yes they do mention they've tested to 4K but in an unreleased version of the model.
      I've read comments on other channels as well and some users are annoyed after downloading and setting it up thinking they're getting 4K images out of the box and that's not the case.

    • @TheFutureThinker
      @TheFutureThinker  29 дней назад

      @@glenyoung1809 this is the most common mistake of people nowadays , see something, zombie brain mode on, then rush to download. Haven't go through the steps or detail. Thats why many said error or something not work, etc. at the final only some are able to use AI and make it work.
      And in the previous video , where I only focus on this AI model research paper and the MIT demo page testing. I did mention the 4K res, and its roadmap.
      Also this is a base model , a lots of people have forgot about this point.

    • @glenyoung1809
      @glenyoung1809 29 дней назад

      @@TheFutureThinker Most people see 4K in the title and that's it, the majority could care less about research papers, only that they can get their hands on the model and start pumping out images.
      The problem is the majority of users forget this is the bleeding edge, all of this is experimental and that comes with being a beta tester(which most users don't realize they are).
      There have been reported issues with blank images and then they conclude this model is crap when in fact this is all experimental and changing rapidly.
      To be fair to users there was that mess with Stable Diffusion 3.5 medium, which was crap and too hastily released and it didn't help that Black Forest released Flux 1.0 which only made SD3.5Med look even worse.

  • @schnauzeprincessin
    @schnauzeprincessin Месяц назад

    Thanks it works

  • @traviswatts1305
    @traviswatts1305 Месяц назад +2

    Did this work for anyone?

    • @vivekkarumudi
      @vivekkarumudi Месяц назад

      i am getting two errors back to back "Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument mat1 in method wrapper_CUDA_addmm" , "Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument mat1 in method wrapper_CUDA_addmm

    • @ДмитрийКарпич
      @ДмитрийКарпич Месяц назад

      After 3 hours of googling and re-installing - yes. But... This model feels like proof of concept, not real thing.

  • @kd4pba
    @kd4pba Месяц назад

    Yea so you lost me at search box when you moved the mouse around.

  • @hebercloward1695
    @hebercloward1695 13 дней назад

    Watched the whole video called "nVidia SANA in ComfyUI - is it worth it?" And at the end they said NOT READY FOR PRIMETIME. So NO you wont be able to get it to work yet as of 1/1/25

  • @SeanieinLombok
    @SeanieinLombok Месяц назад

    brooo waiting on aqua scaping channel now :D

    • @TheFutureThinker
      @TheFutureThinker  Месяц назад

      I know in your place that have beautiful wood 😁 and good water condition for fish

  • @Ryuraaan
    @Ryuraaan Месяц назад

    what's left is to make them read our thoughts xD Typing cost to much time haha

  • @foreropa
    @foreropa Месяц назад

    Nice video, but please, when you do this kind of videos, explain how you do things. I´m searching for a way to search inside ComfyUI and I don´t find anyway to do it, I´m stuck almost at the beginning. Edit: Ok, I found a way and now I don´t get how to do the rest because you don´t explain how to do it, so it´s not a comprehensive tutorial as you call it, it´s a tutorial for people that knows what to do.

  • @chrisgreenwell3404
    @chrisgreenwell3404 Месяц назад

    Now just need a controlnet for it

  • @WiseOwlLearning
    @WiseOwlLearning Месяц назад

    Hope it supports lora, controlnet and ipadapter soon

  • @kalakala4803
    @kalakala4803 Месяц назад

    Haha, i remember our office with a 180cm planted tank 😂 when will you setup a new layout again?

    • @TheFutureThinker
      @TheFutureThinker  Месяц назад

      Maybe coming 2025 I join again the IAPLC 😁🤫. And I need to buy some new stone. Too bad AI cannot generate that. LOL

  • @Kvision25th
    @Kvision25th Месяц назад

    iit works but not sure it bether than sd15 or sdxl :(

  • @petertremblay3725
    @petertremblay3725 Месяц назад

    I have try Sana and i prefer by far flux, something is just not right in the lighting and details.

    • @Gaitchs
      @Gaitchs Месяц назад

      @@petertremblay3725 same with redux and pulid flux still rules

  • @Veselin_Angelov
    @Veselin_Angelov Месяц назад +2

    The question every user wants to ask, and no developer wants to answer:
    "Does it do hands?"

  • @fabiojojo-x9c
    @fabiojojo-x9c 27 дней назад +1

    GemmaLoader prbleme

  • @biggreg100
    @biggreg100 Месяц назад +3

    All I get is a blnk picture on every run.

  • @lucasfrancelino5141
    @lucasfrancelino5141 Месяц назад

    Ai Never sleeps, dam!

  • @youtubeccia9276
    @youtubeccia9276 Месяц назад

    Mindblowing o_O

  • @golddiggerprankz
    @golddiggerprankz Месяц назад

    CPU wahh🤩

    • @TheFutureThinker
      @TheFutureThinker  Месяц назад

      @@golddiggerprankz yup yup the text encoder do the same as T5 but very lightweight.

    • @FusionDeveloper
      @FusionDeveloper Месяц назад +1

      not for generating the image, only for the prompt.

  • @MiracleMan-ol6gd
    @MiracleMan-ol6gd Месяц назад

    The downside with Comfyui is not primarily the cluttered UI, but the difficulty in gauging the settings for an optimal output. Trial and error takes precious time, and there are no standardized and proved settings for the best results. Opinions are like rear ends.

  • @RobertMcDonald_trz
    @RobertMcDonald_trz Месяц назад

    6:55

  • @RikkTheGaijin
    @RikkTheGaijin 2 дня назад

    It's fast but the images look like from 2023

  • @taucalm
    @taucalm Месяц назад +1

    As its about 4 times smaller database than flux it pretty much cant be better and as variable. What comes to VRAM they should publish specific checkpoints. Like people, cats, dogs, buildings, landscapes etc.. 4 sure its faster. So is SDXL.
    Better in quality? Nope. More variance? Nope.
    I hate the hype going around everything released just to get clicks.

  • @purelife_ai
    @purelife_ai Месяц назад

    How about people?

  • @janvollgod7221
    @janvollgod7221 13 дней назад

    fastest black image generator. Somehow the model don't do anything.

  • @eltalismandelafe7531
    @eltalismandelafe7531 Месяц назад +1

    It is wonderful for creating landscapes but for creating humans it is very bad, the worst.

  • @ian2593
    @ian2593 Месяц назад

    Followed exactly but get "`.to` is not supported for `4-bit` or `8-bit` bitsandbytes models. Please use the model as it is, since the model has already been set to the correct devices and casted to the correct `dtype`."

    • @ian2593
      @ian2593 Месяц назад

      updated comfyui today and it's fixed but now I get the black screen. Will try to find that old vae.

  • @peterr6595
    @peterr6595 Месяц назад +1

    But I will let SANA sit for a month while you guys work out the kinks. Too much too fast. BTW 16GB cards will be the minimum for 2025. Buy a new GPU before those orange man tariffs kick in.

  • @emiln1977
    @emiln1977 12 дней назад

    This Sana draws people just terribly :(( No miracle happened, I’m staying with Flux Schnel

    • @koganboss4874
      @koganboss4874 3 дня назад

      Well, what did you expect? It's just a small model of 0.6B parameters, while FLUX is 6B - 11B parameters. Of course the results will be terrible. =)

  • @大支爺
    @大支爺 Месяц назад +2

    Censored.

  • @koganboss4874
    @koganboss4874 3 дня назад

    I didn't see anything surprising, it's just a small model, so it requires less resources and produces terrible results.

  • @comfyuiopenart
    @comfyuiopenart 6 дней назад

    1000 times slower and resalt black frame

  • @glythlablood6295
    @glythlablood6295 6 дней назад

    i need a better computer sadly :/

  • @biggreg100
    @biggreg100 Месяц назад

    I