Flux.1 IMG2IMG + Using LLMs for Prompt Enhancement in ComfyUI!

Поделиться
HTML-код
  • Опубликовано: 14 дек 2024

Комментарии • 155

  • @FlowFidelity
    @FlowFidelity 4 месяца назад +16

    2:41 WHAAAAAT That's how that works! Oh my goodness! I heard someone say the picture is the workflow, but didn't get it. Now I do :) Thank you!

    • @JohnVanderbeck
      @JohnVanderbeck 3 месяца назад +2

      It's one of the magical bits about ComfyUI :)

  • @memoryhero
    @memoryhero 4 месяца назад +17

    The constant quick side commenting was magical in this vid - you kept it brief enough so that veterans won't feel bogged down by old redundant info but also that newbies will highly benefit from it. World class tutorial protocol.

  • @synthoelectro
    @synthoelectro 4 месяца назад +2

    and for those who are stuck with 4GB VRAM, just use a large virtual memory and about 768 x 768, it takes up to 8 mins depending but hey, we did it before on 1.5. and SDXL, we can keep going, you can do this.

    • @riflebird4842
      @riflebird4842 4 месяца назад

      @@synthoelectro what do you mean by use a large virtual memory? Can you explain?

    • @synthoelectro
      @synthoelectro 4 месяца назад

      @@riflebird4842 swap file.

  • @esuvari
    @esuvari 4 месяца назад +54

    Canny for flux has just been released today

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +17

      I thought someone might notice that sneaky screenshot I put in 😉

    • @swannschilling474
      @swannschilling474 4 месяца назад +1

      OMG its really going good!!! 🎉🎉🎉

    • @PhotoBomber
      @PhotoBomber 4 месяца назад +2

      Whats canny?

    • @JustFeral
      @JustFeral 4 месяца назад

      @@PhotoBomber A controlnet model type. Lines and such, google it.

    • @256chiru
      @256chiru 4 месяца назад

      What is canny

  • @Cadmeus
    @Cadmeus 4 месяца назад +2

    A node that really helps to manage latent/image sizing btw is an underappreciated little extension from Ser-Hilary, called SDXL_sizing. Automatically spits out the right size for any base resolution (e.g. 512, 1024, 2048), at any given aspect ratio. I use wildcards from Impact Pack as an input to the aspect ratio setting, which works *really* well with Flux.

  • @Copperpot5
    @Copperpot5 4 месяца назад +1

    Excellent job on this workflow! Playing w/ it now after making a few of my own/using some common on civ/discord, but your incorporation of the LLM Party node + autosizing/etc is simply brilliant. Hope all is well!

  • @akratlapidus2390
    @akratlapidus2390 4 месяца назад +4

    Nerdy, you always deliver!!! 👌🏻👏🏻👏🏻👏🏻👏🏻

  • @BirkB1
    @BirkB1 4 месяца назад +2

    Thank you. LLM party sounds awsome 😄

  • @paulotarso4483
    @paulotarso4483 2 месяца назад

    love this narration haha thank you!

  • @jonmichaelgalindo
    @jonmichaelgalindo 4 месяца назад +1

    Even if outputs were non-commercial, education and reporting are protected fair use. (Education and reporting are inherently commercial. Teachers and journos have to get paid.) Fair use is the doctrine all AI model training is built on. 😊

  • @purposefully.verbose
    @purposefully.verbose 4 месяца назад +8

    "nice beaver"
    ok, thanks for that.

    • @joeduffy52
      @joeduffy52 4 месяца назад +1

      I've just had it stuffed.

  • @massibob2004
    @massibob2004 3 месяца назад +1

    My god ! You re good :)

  • @MilesBellas
    @MilesBellas 6 дней назад

    "Pandora's Box of Oddities!"
    😅🤣👍

  • @DeconvertedMan
    @DeconvertedMan 4 месяца назад +6

    :) cute things AI makes are cute.

  • @wakegary
    @wakegary 4 месяца назад

    Bill didn't seem to mind the multiples of 16! Stay nerdy!

  • @deadlymarmoset2074
    @deadlymarmoset2074 4 месяца назад +7

    OOOOOooh NErdy rODEnt...

  • @erikjohnson9112
    @erikjohnson9112 4 месяца назад

    Almost 50K subs. I'll do my part. (just subbed)

  • @icchansan
    @icchansan 4 месяца назад +1

    Amazing, tutorials to make a custom lora soon?

  • @blakecasimir
    @blakecasimir 4 месяца назад +1

    I hope FOOOCUS adds support for Flux

  • @joeb2920
    @joeb2920 4 месяца назад +1

    Do you have a copy of the final workflow that we can download?

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      Of course! www.patreon.com/posts/ai-enhanced-flux-109665789

  • @raducodreanu2309
    @raducodreanu2309 4 месяца назад

    Hi, thanks for the awesome tutorial! How did you get the labels updated for the switches input (ex: "1_llm_enhanced", instead of "text1")?

    • @joeduffy52
      @joeduffy52 4 месяца назад

      Right-click on the small dot next to the label and you should see "Rename Slot".

  • @scarletsword45
    @scarletsword45 4 месяца назад +4

    I really like Flux, but I'm disappointed that I keep having to buy a new GPU to keep up with the demands of new AI art models. 😁

    • @SouthbayCreations
      @SouthbayCreations 4 месяца назад +3

      Here’s my theory, buy the biggest (consumer) gpu available, 4090, and be good for a few years. People will spend the minimum and expect to be good forever. In the AI world models are only going to keep getting bigger and more reliant on vram. It’s unfortunate but if we want better AI then it comes with a cost

  • @dagkjetsa8486
    @dagkjetsa8486 4 месяца назад

    Hello. Great tutorial! :) Though... I am trying to follow your tutorial and get llama 3.1 up and running in comfyui, but it is not working. I am not sure what to put into the base_url field of the API Large Language Model Loader. Any help? Also, why are you using this node and not the "Local Large Language Model"?
    Thanks

  • @equilibrium964
    @equilibrium964 4 месяца назад

    A face detailer workflow for flux would be really useful.

  • @danowarkills4093
    @danowarkills4093 29 дней назад

    Can we get the img2img workflow?

  • @eveekiviblog7361
    @eveekiviblog7361 3 месяца назад

    it says to install flash attention. where to get it and where to put?

  • @michaelbayes802
    @michaelbayes802 4 месяца назад +1

    The ability to load the flux models via "Load Checkpoint" thows up an error for me ...Could not detect model type of ... did you have this problem?

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      Are you using the special fp8 checkpoint downloaded via the link above workflow, like in the video?

  • @yngeneer
    @yngeneer 4 месяца назад +1

    Lets Party 🎉🥳

  • @TobinatorXXL
    @TobinatorXXL 4 месяца назад

    hello my LLM party problem: Error code: 500 - {'error': {'message': 'llama runner process no longer running: -1 ', 'type': 'api_error', 'param': None, 'code': None}}

  • @magimyster
    @magimyster 4 месяца назад +3

    goodbye mj🤭👍

  • @tosvus
    @tosvus 4 месяца назад

    Does your patreon have workflows that work with the largest flux dev model? I have the regular one working fine, but it looks quite different from what is here, (and uses the 24GB file), and I would like to get all the extra stuff on your patreon if it works with that as a basis (or basically works with 24GB related files). BTW: I use a RXT 4090. Thanks!

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      Yup, there’s a whole boatload of workflows 😉

    • @tosvus
      @tosvus 4 месяца назад

      @@NerdyRodent Thanks, signing up!

  • @carlodemichelis
    @carlodemichelis 4 месяца назад

    How do you add/remove Pin(s) to slots of a node? For example, the pin to the t5xxl slot (wich become hidden) in the CLIPTextEncodeFlux node, or the user_prompt pin in the API Large Language Model? (sorry, complete noob in Comfy). Thanks.

    • @NerdyRodent
      @NerdyRodent  4 месяца назад

      You can write click on any node for a variety of options,

    • @carlodemichelis
      @carlodemichelis 4 месяца назад

      @@NerdyRodent Yup, of course, but i don't find any options to add/remove pins to slots :-/

  • @blakemann2794
    @blakemann2794 4 месяца назад

    I don't have comfy installed yet... but I've been wanting to try flux img2img with my DAZ3D renders... I'd be interesting to see how subtle I can make the changes... I just want to make the renders look like either painted/illustrated or more realistic.. while keeping as much detail as possible.. I got some good results a1111 by simply upscaling renders with certain models and prompts in effect... I'm wondering if I can do the same with Flux

    • @TR-707
      @TR-707 4 месяца назад

      install it already

  • @DrMacabre
    @DrMacabre 4 месяца назад

    is there a way to reduce the memory usage of Llama ? it's peaking heavily on my 3090 and makes everything painfully slow.

    • @NerdyRodent
      @NerdyRodent  4 месяца назад

      Using fp8 and a smaller LLM should reduce the load. With 64gb RAM and a 3090 you should see images generate in around 20 seconds

    • @DrMacabre
      @DrMacabre 4 месяца назад

      @@NerdyRodent damn, im already on fp8 with the smallest Llama 3.1 model, it goes just enough over the vram to slow everything down with Florence + Llama.

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      you can change the amount of time all models are loaded into memory by setting the `OLLAMA_KEEP_ALIVE` environment variable when starting the Ollama server. You'll need to wait for it to load again each time, but if you've got other things eating that VRAM it may help

    • @DrMacabre
      @DrMacabre 4 месяца назад

      @@NerdyRodent Done, no more long queue, thank you :) the results with florence and Llama are simply amazing

    • @NerdyRodent
      @NerdyRodent  4 месяца назад

      @@DrMacabre yeah, it totally changes how you can prompt!

  • @MediAndLemon
    @MediAndLemon 4 месяца назад

    Do Loras work yet (and I am just too stupid for it) or are those also still on the waiting list?

    • @NerdyRodent
      @NerdyRodent  4 месяца назад

      Yup! Loras and controlnets are available, though still early days!

    • @Utoko
      @Utoko 4 месяца назад

      yes but of course they need additional VRam controlnet too

  • @shareeftaylor3680
    @shareeftaylor3680 4 месяца назад

    Can you please compare the different flux gguf versions please

  • @jjog3185
    @jjog3185 4 месяца назад +1

    Is it possible to run it locally with an RTX4060 8GB VRAM and with 16GB RAM ?

    • @BlackParade01
      @BlackParade01 4 месяца назад +1

      @mexihcahcoatl4105 that's with Schnell, right?

    • @jjog3185
      @jjog3185 4 месяца назад

      @mexihcahcoatl4105 I'll try and I'll share my result and opinion. Thanks!

    • @kkryptokayden4653
      @kkryptokayden4653 4 месяца назад

      @@BlackParade01 I would use dev version is way better, slower but is worth it

    • @BlackParade01
      @BlackParade01 4 месяца назад

      @@kkryptokayden4653 I've used both and tested both intensively. The Dev version is really good for photorealism, but the Schnell model seems great for illustrated outputs. And of course, it's much faster.
      Both models have their uses. I prefer the Dev, but I use the Schnell for img2img.

    • @luislozano2896
      @luislozano2896 4 месяца назад +2

      Even having 32 GB of ram it will max fill and swap memory and slow down! Close all other apps and tabs! i just got some extra ram yesterday to 48 gb. i just found the FP8 version and got a bit of speedup.

  • @Kvision25th
    @Kvision25th 3 месяца назад

    Im starting with flux can you share that workflow??

    • @NerdyRodent
      @NerdyRodent  3 месяца назад

      Sure! The pre-made version you can grab from www.patreon.com/posts/ai-enhanced-flux-109665789

  • @MilesBellas
    @MilesBellas 2 месяца назад

    Llama 3.2 for Comfyui FLUX ?
    Prompt enhancement node?

    • @NerdyRodent
      @NerdyRodent  2 месяца назад +1

      Yup, works great with llama3.2… or indeed much larger models

  • @JohnVanderbeck
    @JohnVanderbeck 3 месяца назад

    Can someone help me understand why the fp16 model has to be used from unet and the fp8 can just be used as a normal checkpoint? What even is the difference between unet and checkpoints and why does it matter for what is essentially the same model?

    • @NerdyRodent
      @NerdyRodent  3 месяца назад

      Think of the collection of files like a zip file, where lots of things are collected together - whereas the unet is just one of those files.

  • @Tapiolla
    @Tapiolla 4 месяца назад

    Excuse me, but I didn't understand, where can I find the workflow you have shown in the video?

    • @SouthbayCreations
      @SouthbayCreations 4 месяца назад +1

      On his patreon page

    • @Tapiolla
      @Tapiolla 4 месяца назад

      @@SouthbayCreations Thank you! Wasn't clear :)

  • @CapaUno1322
    @CapaUno1322 4 месяца назад

    Hi buddy, what GPU are you using? I have an rx6800 which I am really happy with and it's 16gb for half the price of Nvidia as of last week and hopefully I will be able to get it to work with a few tricks here and there....just wondering...thanks! ;D

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      I’ve got an old 3090 as VRAM is where it’s at. AMD cards should work for the most part on Linux, though on MS Windows it’s likely to be a slightly more bumpy journey and may not work at all for many things!

    • @CapaUno1322
      @CapaUno1322 4 месяца назад

      @@NerdyRodent Thanks for your reply, I've had a few 'bumps' for Stable D but I did overcome then, I've discovered ZLUDA and Anaconda and peeps have things up and running but it's a lottery to how much ball ache you may have to experience, one guy is getting really fast renders with ZLUDA on AMD so eh, fingers crossed, thanks for your helpful videos and I'll let you know, just to add that the rx6800 with gaming is only 20% lower on average with FPS than the rtx3090, I know there's more to AI but that's really good and a used 3090 is still $6-700 depending on which one...so hopefully I'll get some reasonable performance as well....have a fun day!

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      @@CapaUno1322 nice!

  • @themachine8229
    @themachine8229 4 месяца назад

    Works with flux1 dev.sft?Because I don't have the extension.safetensor in the model I use

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      Yup! Remember to use the workflows at the top when using the individual files though

    • @themachine8229
      @themachine8229 4 месяца назад

      @@NerdyRodent thanks bro

  • @squallseeker-i2i
    @squallseeker-i2i 4 месяца назад

    Not mentioned, but I presume that to use the standard checkpoint loader I must move the flux models from unet to where they will be visible to the std loader. I ran the update-all before getting started tonight and flux is basically broken on my 3080 now after using it all week... so I don't have a choice other than trying to modify the workflow.

    • @simonmunk4326
      @simonmunk4326 4 месяца назад +2

      @@squallseeker-i2i The checkpoint version of the model is not the same file as the unet version. Move the unet version back and download the checkpoint version. The checkpoint version is around 18 GB.

    • @NerdyRodent
      @NerdyRodent  4 месяца назад

      @squallseeker. No - use the linked files above each workflow as they’re appropriate to that workflow. For example, to use the one shown in this video you’ll need to download the file as shown, directly to the location shown.

  • @엠케이-p3p
    @엠케이-p3p 4 месяца назад

    who are the people that can run this workflow? I am on 12gb 3060 gpu, is it possible to run flux and ollama at the same time? does anyone know about this?

  • @EllaFinch0812
    @EllaFinch0812 4 месяца назад

    Hi, do you accept sponsorship to your video?

  • @sinayagubi8805
    @sinayagubi8805 4 месяца назад

    can you somehow simulate negative prompts??

    • @tc8557
      @tc8557 4 месяца назад +1

      @@sinayagubi8805 since it's an llm you're prompting, just say what you dont want.
      Try like 'with no eyeglasses or makeup on'

    • @Elwaves2925
      @Elwaves2925 4 месяца назад

      From what I've read, Flux doesn't use negative prompts. If you add one it will often ignore it.
      Try what @tc8557 says and describe what you don't want.

  • @EmmaFitzgerald-dp4re
    @EmmaFitzgerald-dp4re 4 месяца назад

    thanks for the vid, really appreciate it!, But the resources needed to run flux, It's still too just demanding for me

  • @juanjesusligero391
    @juanjesusligero391 4 месяца назад +3

    Oh, Nerdy Rodent, 🐭
    he really makes my day; 😎
    showing us AI, 🤖
    in a really British way. 🫖
    🎵🎶

  • @TR-707
    @TR-707 4 месяца назад

    hmm whats this ays + scheduler. Also, is Flash attention 2 required for maximum nerdyness or not really?

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      There are multiple options for florence, but I find flash attention works just fine! Align Your Steps is just an option that works quite well - ruclips.net/video/JgYzCEzHDrc/видео.html

    • @TR-707
      @TR-707 4 месяца назад +1

      @@NerdyRodent wow awesome video! I don't see AYS+ in my ksampler though. I do see Beta and Resample though ..

    • @Rentoa
      @Rentoa 4 месяца назад

      i have the same problem... no ays+ scheduler?

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      @@TR-707 Align Your Steps is just an option that works quite well - ruclips.net/video/JgYzCEzHDrc/видео.html

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      @@Rentoa Align Your Steps is just an option that works quite well - ruclips.net/video/JgYzCEzHDrc/видео.html

  • @godfuzza2778
    @godfuzza2778 3 месяца назад

    Has anyone shared this workflow already or do I have to build it on my own?

    • @NerdyRodent
      @NerdyRodent  3 месяца назад +1

      You can get the workflows here - www.patreon.com/posts/ai-enhanced-flux-109665789

  • @CheoWalker
    @CheoWalker 3 месяца назад

    CLIPTextEncodeFlux as of this day doesn't have the t5xxl input

  • @choppergirl
    @choppergirl 4 месяца назад +2

    I found ComfyUI impossible to figure out.

    • @shApYT
      @shApYT 4 месяца назад +3

      Then you'll have trouble using any other art software. Nodes aren't hard. Just do it. blender, houdini, substance designer, davinci resolve etc.

    • @choppergirl
      @choppergirl 4 месяца назад

      @@shApYT Funny. I tried to get the demo sample project to do anything at all with no luck.
      I use othe AI programs all the time. ComfyUI just looked like a logic flowchart mess of object oriented containers gone off the rails.

    • @shApYT
      @shApYT 4 месяца назад +1

      @@choppergirl millions of artists using blender, Houdini, grasshopper, davinci resolve, substance and countless. It is the standard interface.

    • @goodie2shoes
      @goodie2shoes 4 месяца назад

      did you try the portable /standalone version? and what are your GPU specs?

    • @Elwaves2925
      @Elwaves2925 4 месяца назад

      @@shApYT That's complete rubbish about other art software.

  • @NotThatOlivia
    @NotThatOlivia 4 месяца назад +1

    this is not the best approach - why load into RAM/VRAM LLMS and corresponding nodes, if you can create prompts with them and after run comfy to generate?

  • @robbana9909
    @robbana9909 4 месяца назад

    Generating images with the checkpoint models takes forever on my 8gb card compared to the unet models for some reason.

    • @luislozano2896
      @luislozano2896 4 месяца назад

      this new checkpoint is huge! the regular Flux one is 22GB, the FP8 version is 11GB. We got so used to having SD1.5 at 2-4 gb! XL and pony 6GB

  • @FusionDeveloper
    @FusionDeveloper 4 месяца назад

    My version is:
    2489 (comfyui)
    2.48.5 (manager)

  • @riflebird4842
    @riflebird4842 4 месяца назад

    What 4GB vram people can do ??😢

    • @NerdyRodent
      @NerdyRodent  4 месяца назад

      Sd1.5 is a great choice for extra low vram cards!

  • @elgodric
    @elgodric 4 месяца назад

    Of all of the 8 billion photos out there you chose Bill fuckin Clinton 👌

  • @SlyNine
    @SlyNine 4 месяца назад +1

    How do I get the manager menu to come up. I'm ew to ComfyUI, sorry if it's a dumb question

    • @NerdyRodent
      @NerdyRodent  4 месяца назад +1

      ComfyUI manager is the first thing to install right after Comfy itself - github.com/ltdrdata/ComfyUI-Manager

  • @Silberschweifer
    @Silberschweifer 3 месяца назад

    the moment you don't check he mean schnell > Schn-e-ll

  • @fast_harmonic_psychedelic
    @fast_harmonic_psychedelic 4 месяца назад

    If it uses CLIP why dont you just use clip vision encoder to input an image and trick flux by telling it its a text embedding lol

  • @vitalis
    @vitalis 4 месяца назад +1

    I see rodent I click. Simple 👍

  • @hadbildiren123
    @hadbildiren123 4 месяца назад +1

    Not a nice model. Generations look like animation or drawing and the skin is like plastic or shiny! Going on with SDXL!

    • @daylight3d
      @daylight3d 4 месяца назад +1

      Don't use the Schnell version. It's plastic. Dev version is much better.

    • @kkryptokayden4653
      @kkryptokayden4653 4 месяца назад

      @@hadbildiren123 I agree the dev version and the schnell version are very different it's day and night

  • @ArchangelAries
    @ArchangelAries 4 месяца назад +2

    I hate comfyui with a visceral passion

    • @generichuman_
      @generichuman_ 4 месяца назад

      Oh, so you don't know how to use it and you've never even really tried... thanks for sharing!

  • @_.o..o._
    @_.o..o._ 4 месяца назад

    I'm usually very sceptical about "these" things, but why do most of the major RUclips channels with AI tutorials always use a political figure to explain image and video generation with AI? I mean, come on 🤐

    • @Elwaves2925
      @Elwaves2925 4 месяца назад +1

      I'd say it's because it's someone recognisable, a public figure of their own choosing, that doesn't fall into the category of actors etc who don't. Or it's something else, like them being easy targets.

    • @_.o..o._
      @_.o..o._ 4 месяца назад

      ​@@Elwaves2925 I would rather see actors' faces. Lately, whatever I try to learn, I see political figures shoved down my throat everywhere 🤮

    • @Elwaves2925
      @Elwaves2925 4 месяца назад

      @@_.o..o._ I don't think it's being done because it's political, not on this channel anyway but for me using actors should be okay as long as nothing outrageous is done with them. 🙂

  • @qus123
    @qus123 4 месяца назад

    I tried loading flux with the same loading checkpoint as you. It fails with: Error occurred when executing CheckpointLoaderSimple:
    ERROR: Could not detect model type of: /mnt/e/Projekty/_AI/ComfyUI/models/checkpoints/Flux/flux1-dev-fp8.safetensors
    (this model does load the "old" way, though)

    • @NerdyRodent
      @NerdyRodent  4 месяца назад

      Could be the checkpoint file or your version of comfy? Make sure it’s the 17gb one and check the sha256