CivArchive
    Windtunnel Face - Flux.1-Dev
    NSFW
    Preview 50959438
    Preview 50959483
    Preview 50959721
    Preview 50960046
    Preview 50960315
    Preview 50960591
    Preview 50960719
    Preview 50960849
    Preview 50961017
    Preview 50962206
    Preview 50962836
    Preview 50963295
    Preview 50998046
    Preview 51032042
    Preview 51060883
    Preview 51227070
    Preview 51227250
    Preview 51228359
    Preview 51363133

    This LoRA makes images of people who look like they're in a wind tunnel. I wanted this for some reason, so here it is lol.

    Trigger is w1nd8l0wn photo, but it helps to add things like closeup, windy, and/or wide open mouth and flapping cheeks.

    NOTE: The Z-Image Turbo version doesn't seem to play nice with character LoRAs. It can distort the likeness and introduce errors/deformations in the scene. Try using more (e.g., 18) steps and increasing the model shift (e.g., 10). Also, for less noisy/blotchy images use the FlowMatch Euler Discrete Scheduler along with Euler_ancestral.

    If you can't get a closeup photo, make sure you aren't describing things like pants, shoes, excessive background details, etc.

    Description

    FAQ

    Comments (21)

    TigonTXJan 11, 2025· 3 reactions
    CivitAI

    Lol, this is amazing, thx!

    bblink787
    Author
    Jan 11, 2025

    Thanks! :D

    bblink787
    Author
    Jan 11, 2025· 1 reaction

    And thanks for the Buzz!

    BuffaloSymphonyJan 11, 2025· 6 reactions
    CivitAI

    I'm dying this is such a great concept

    bblink787
    Author
    Jan 11, 2025

    Thanks! I'm still cracking myself up here :D

    peopleofpens939Jan 11, 2025
    CivitAI

    Loving it so far, What strength do you recommend, also how do you get your images so sharp and high resolution?

    bblink787
    Author
    Jan 11, 2025· 1 reaction

    Thanks! I've been running it at a strength of 1.0 and it's been working pretty well without introducing weird deformities (ok, weirder than it's supposed to be!). I'm using a pretty basic workflow in Comfy (each image has the workflow in it, if you're a Comfy user). I'm running full beans with flex1-dev.safetensors at fp16 (that's probably the biggest thing I do for image quality), and no model sampling. Euler/Normal for 26 steps at a guidance of 3.0. I guess I tend to use the word "vibrant" a lot to get a little more saturation too.

    peopleofpens939Jan 11, 2025· 1 reaction

    @bblink787 Thats great! thanks for the reply, i shall try those settings!

    PolygonJan 11, 2025· 6 reactions
    CivitAI

    OMG this is really funny!
    Whoever said AI can't do art clearly hasn't seen this yet 🤣🤣🤣

    hamstornadoJan 15, 2025· 2 reactions
    CivitAI

    This LORA... this is the real deal. such and wildely impressive piece of work, well done to you, and thank you for sharing!

    bblink787
    Author
    Jan 15, 2025

    Thank you for the kind words! I did several trainings and loads of testing to find the sweet spot. It's great to be appreciated! :D

    101033Feb 2, 2025· 4 reactions
    CivitAI

    Combining this with the Black Hole Sun Lora would be.... something

    slimedFeb 10, 2025

    you graduated at cornell huh?

    tumbleworld365Feb 7, 2025
    CivitAI

    Enough with the sock puppet spam, dude. It's getting seriously pathetic.

    EDIT: Why did I assume it was OP spamming this LoRA? Naive of me. Rude, too.

    bblink787
    Author
    Feb 8, 2025

    Lol, wrong model, but I totally feel your pain 😂

    bblink787
    Author
    Feb 8, 2025

    Oh shit, wait, is someone posting sock puppet crap under my account??

    bblink787
    Author
    Feb 9, 2025

    I reported this btw. I'm not the only one this is happening to. I'm removing them from the gallery one by one, but it's taking forever.

    tumbleworld365Feb 9, 2025

    @bblink787 I'm sorry for assuming it was you that had gone insane. I did seriously wonder why... I dunno, stressful times is my only excuse. I can't imagine how much work it is to delete them one by one. Thanks for letting me know.

    bblink787
    Author
    Feb 9, 2025

    @tumbleworld365 It's all good. I checked on the mod-report channel on Civitai's discord and apparently this is getting more common. I don't get how anyone benefits though? Anyway, I put in a ticket asking for a bulk-remove feature ;)

    tumbleworld365Feb 10, 2025

    @bblink787 Yeah, I've been wracking my brain, but I'm damned if I see what the benefit is. Testing the bot-tech is about the best I've got.

    jaydenny28th725Mar 1, 2025
    CivitAI

    can you upload to huggingface if thats possible?

    LORA
    Flux.1 D

    Details

    Downloads
    628
    Platform
    CivitAI
    Platform Status
    Available
    Created
    1/11/2025
    Updated
    8/14/2026
    Deleted
    -
    Trigger Words:
    w1nd8l0wn photo

    Files

    w1nd8l0wn photo_FLUX.safetensors

    Mirrors