sorted by: new top controversial old
If anyone wants to join perchance would have to be now in c/perchance@lemmy.world
If anyone wants to join perchance would have to be now in c/perchance@lemmy.world
If anyone wants to join perchance would have to be now in c/perchance@lemmy.world
Where is the "negative prompt" in the ai-text-to-image-generator? in c/perchance@lemmy.world
[–] 2 points 7 months ago

Thats true! Since nat lang text encoders are more complex , the negatives of the stuff being encoded is rarely the opposite.

As in , the model itself (Chroma on perchance) isnt trained to comprehend negative vectors. Tho lodestones (creator of Chroma) never specified but I assume not.

Negatives are an off shoot in training that sorta worked in CLIP based models SD1.5 SDXL (Pony / illustrious) that carried over into the nat lang mod releases out of habit in the community.

It worked in CLIP models because the CLIP encoder is simple. Write 'ice cream' in CLIP and the text encoding vector will point roughly same direction no matter where in the 75 token batch 'ice cream' .

Compare that to the many number of different answers you can get from chatGPT or GROK containing the word 'ice cream' and you can see how the 512 token batch encoding of the T5 in Chroma , or Qwen encoder in Klein / Z image varies drastically depending how common words are arranged in the text.

  • permalink
  • fedilink
  • source
  • parent
  • context
    • (Content Warning) Threat message in c/perchance@lemmy.world
    [–] 3 points 1 year ago

    Perchance chat is gonna get the site shut down I swear.

    Any forum with a buncha unhinged weirdos + underage users is trouble waiting to happen for the site. People wanna chat , do that on Lemmy or whatever other moderated forums there are IMO.

  • permalink
  • fedilink
  • source
    •
    Report: Underage User in 18+ General Chat in c/perchance@lemmy.world
    Report: Underage User in 18+ General Chat in c/perchance@lemmy.world
    Report: Underage User in 18+ General Chat in c/perchance@lemmy.world
    help creating a var or style for camera + lens size&type&focal length + shot type&shot angle, then maybe after, different filters? in c/perchance@lemmy.world
    Trying to recreate "casual photo" style locally in c/perchance@lemmy.world
    Trying to recreate "casual photo" style locally in c/perchance@lemmy.world
    Trying to recreate "casual photo" style locally in c/perchance@lemmy.world
    [–] 1 point 1 year ago* (last edited 1 year ago) (4 children)

    Perchance model is FLUX Chroma.

    FLUX Chroma Flash Heun

    Easiest way to get photoreal output is to head off to getty and copy a photo caption: https://www.gettyimages.com/editorial-images

    Example output on perchance generator

    2025 Toronto International Film Festival - Black Excellence Brunch TORONTO, ONTARIO - SEPTEMBER 08: (L-R) Karen Chapman, Regina Taylor and Joan Jenkinson attend the Black Excellence Brunch during the 2025 Toronto International Film Festival at Petros82 Restaurant on September 08, 2025 in Toronto, Ontario. (Photo by Leon Bennett/Getty Images)

    From : https://www.gettyimages.com/editorial-images/entertainment/event/toronto-international-film-festival-black-excellence-brunch/776375508?editorialproducts=all

  • permalink
  • fedilink
  • source
    •
    0
    Train your own custom image captioning model for Chroma (Perchance T2i model) on Google Colab T4 in 2 hours (huggingface.co)
     

    Link to image-to-prompt: https://huggingface.co/codeShare/flux_chroma_image_captioner/blob/main/gemma_image_captioner.ipynb

    Writing prompts for Chroma is hard and Joycaptions is inaccurate so I assembled the training data I could find for the model , picked 400 image text pairs at random and trained Google Gemma 3 LoRA model as in image to prompt tool that can run on Google Colab.

    Its a proof-of-concept. Feel free to train your own LoRa captioning models for use on perchance. The workflow of converting JSON and .parquets into a dataset can be found in this notebook in the repo: https://huggingface.co/codeShare/flux_chroma_image_captioner/blob/main/train_on_parquet.ipynb

    For the original unsloth notebook visit: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Gemma3_(4B)-Vision.ipynb

    Other unsloth models: https://docs.unsloth.ai/get-started/unsloth-notebooks

    /Cheers!

    1
    Train your own custom image captioning model for Chroma (Perchance T2i model) on Google Colab T4 in 2 hours (huggingface.co)
     

    Link to image-to-prompt: https://huggingface.co/codeShare/flux_chroma_image_captioner/blob/main/gemma_image_captioner.ipynb

    Writing prompts for Chroma is hard and Joycaptions is inaccurate so I assembled the training data I could find for the model , picked 400 image text pairs at random and trained Google Gemma 3 LoRA model as in image to prompt tool that can run on Google Colab.

    Its a proof-of-concept. Feel free to train your own LoRa captioning models for use on perchance. The workflow of converting JSON and .parquets into a dataset can be found in this notebook in the repo: https://huggingface.co/codeShare/flux_chroma_image_captioner/blob/main/train_on_parquet.ipynb

    For the original unsloth notebook visit: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Gemma3_(4B)-Vision.ipynb

    Other unsloth models: https://docs.unsloth.ai/get-started/unsloth-notebooks

    Also Tensor Art holds a contenst for running the new Qwen model so you might wanna check that out: https://mee6.xyz/i/vSpIL2tvi0

    /Cheers!

    7
    All images used to train Chroma (Perchance T2i model) are tagged with the 'aesthetic' tag (lemmy.world)
     

    Source: https://huggingface.co/lodestones/Chroma/discussions/72

    Chroma (Perchance Text to image model) is trained on 5 million images

    The tagging system for all these images include the word 'aesthetic' in the training prompt , used in this manner :

    'aesthetic 0' , 'aesthetic 1' , .... 'aesthetic 10' , 'aesthetic 11' are labels used to denote the visual style in the training data

    Where 'aesthetic 11' denotes (good) AI-images used for training data

    Thats all we know.

    //----//

    This system isn't 100% accurate , but it is highly recommended you use the term 'aesthetic' at least once (preferably often) to mimic the training prompt(s) in the Chroma model.

    Check the HF page for Chroma in the future on further info regarding training data / prompts you can use while generating images on the perchance website.

    TLDR; use the word 'aesthetic' in your prompt to improve them for perchance text-to-image generation

    Cheers!

    7
    FLUX Chroma (The Perchance T2i model) (huggingface.co)
     

    FLUX Chroma: https://huggingface.co/lodestones/Chroma

    FLUX Chroma (Tensor Art) : https://tensor.art/models/886764918794154122

    Unlike base FLUX Schnell , FLUX Chroma uses NAG (Normalized Attentive Guidence) : https://huggingface.co/spaces/ChenDY/NAG_FLUX.1-dev

    TLDR; NAG are negatives added to FLUX model

    Paper: https://arxiv.org/abs/2505.21179

    See FLUX Chroma Huggingface repo for additional changes from base FLUX Schnell model

    To help with creating prompts for FLUX , use Joycaptions:https://huggingface.co/spaces/fancyfeast/joy-caption-beta-one

    And Danbooru tags: https://donmai.moe/wiki_pages/help:home

    The prompt can be up to 512 tokens long , which can be checked at https://sd-tokenizer.rocker.boo/

    view more: next ›