CivArchive
    Preview undefined
    Preview undefined

    AndroFlux is a LoRA model made for FLUX.1 dev by Black Forest Labs, because there’s a glaring lack of accurate male anatomy in AI models. I’m trying to fix that by making sure AI can depict male features more accurately, balancing out the overwhelming female-centric content you see everywhere.

    Community and Support

    Join The Bulge Discord at https://thebulge.xyz if you’re into gay AI art—it’s a great place to share and learn. If you want to support the project, whether with feedback or donations, it’s much appreciated.

    Current Status

    The model isn’t perfect, so don’t expect flawless results.

    Purpose

    AndroFlux is all about making sure AI models can represent male anatomy—especially penises—accurately. With that said, it also has a general photographic aesthetic and focuses on twinks so it is still useful for other SFW purposes. It’s a step towards more balanced AI-generated art.

    Technical Details

    AndroFlux was built and tested primarily on FLUX.1 dev, but it should also be compatible with the Schnell base models, including the quantized fp8 version. These are distilled models, making training tricky and affecting the LoRA’s performance, requiring a lot of optimization.

    Training Process

    • Resolution Journey: I started training at 1024px, but the model wasn’t converging, so I dropped to 512px. The quality wasn’t great, so I ended up at 768px. This resolution offers a good balance between detail and training time.

    • Captioning: Captioning was crucial in training. Contributions came from the AndroAtlas dataset, part of the AndroFlow project. Ephemeron01 from The Bulge Discord assisted with captions using models like llava-hf/llava-v1.6-mistral-7b-hf and a custom-trained PaliGemma VLM. Specific prompts like "man, penis" and the activator "AndroFlux" were also used in later versions.

    • Acknowledgments: Huge thanks to bghira and Terminus Research Group for their support, including providing an A100 for some of the training runs.

    Usage Tips

    • For best results, use prompts that specifically mention male frontal nudity, such as "penis."

    • Recommended settings:

      • Resolution: 1024x1024 or 832x1216

      • Flux Guidance: 3.5

      • LoRA Weight: Start at 1.0, reduce to 0.5 if artifacts appear

      • CFG: 2.0 (1.0 will be faster though)

    Suggested Prompts

    • A nude man in his {late teens | early twenties | early thirties}

    • His penis is {erect | flaccid}

    • His penis is {large | small | average}-sized

    Version History

    See "About this version" on the right for version-specific updates.

    Description

    Note: This version still sucks. I don't really know what I'm doing wrong.


    Training changes from the previously posted run (v5):

    • Training Steps: Maximum of 6000 >> 10000

    • Base Model Precision: int8-quanto >> fp8-quanto

    • Learning Rate: 1.0

    • Optimizer: Prodigy

    • Batch Size: 1 >> 2

    • Gradient Accumulation Steps: 1

    • Rank: 16 >> 8

    • Gradient clipping: default (probably 2) >> 0.1

    • Caption Strategy: Triple >> Instance prompt

    FAQ

    Comments (5)

    vommieAug 9, 2024· 2 reactions
    CivitAI

    Thanks, appreciating your effort, I'm looking forward for updates!

    JesseAug 9, 2024· 2 reactions
    CivitAI

    It's awesome to see you working on this traditionally underrepresented subject in AI image generation, thanks! What does it take in terms of hardware and compute hours to start achieving a good LoRA?

    TheGreatOne321Aug 10, 2024· 2 reactions
    CivitAI

    I know you've said the latest iteration "still sucks" but it's still a big improvement over the earlier attempts!

    As for recommendations, I would suggest checking out the x-flux GitHub page if you're still using SimpleTuner as people seem to be having more success with the training scripts they've provided. I've also noticed that the LoRAs trained with Adam tend to turn out better than those using Prodigy.

    Lastly, I've heard a few people bring up that DoRA should be easier to train and should produce better results, but I'm not sure if the x-flux script supports it, not sure if Comfy supports it, and haven't seen many actual examples of it's use. Still might be worth looking into though?

    Best of luck!

    TheGreatOne321Aug 10, 2024· 1 reaction

    Just thought I would update that a major breakthrough might have been made:

    https://gist.github.com/AmericanPresidentJimmyCarter/0134fc7848aac7025d0c967c6c4df53b

    NuxSinistra
    Author
    Aug 11, 2024· 2 reactions

    @TheGreatOne321 I love you <3

    LORA
    Flux.1 D

    Details

    Downloads
    172
    Platform
    CivitAI
    Platform Status
    Available
    Created
    8/9/2024
    Updated
    4/30/2026
    Deleted
    -
    Trigger Words:
    man
    penis

    Files

    AndroFlux-v8-10k.safetensors

    Mirrors

    Huggingface (1 mirrors)
    CivitAI (1 mirrors)

    Available On (1 platform)

    Same model published on other platforms. May have additional downloads or version variants.