modelos/Chroma - v1.0-HD

Chroma - v1.0-HD

9/7/2025

10:52:11 PM

Palabras Clave y Etiquetas Relacionadas

5m dataset training,8.9b parameter model,apache 2.0 licensed,modelo base,base_model,checkpoint,chroma,chroma stable diffusion model,flux.1-schnell,lodestone,open-source ai,other,v1.0-hd

Close-up of a menacing humanoid creature with gray decaying skin, red glowing eyes, sharp teeth, and pointed ears crouching in a forest with blurred green foliage background.

Black and white photograph of a weathered concrete wall with an oval black hole and a rusted metal door slightly ajar, surrounded by dry grass and bare branches.

A haunting figure with a distorted screaming face wrapped in a translucent cloth and bandaged head, enveloped in swirling smoke from a left side view.

A pale, segmented alien creature with curved bone-like protrusions and tentacle appendages lies on a wet metallic deck under a foggy sky, next to a person in dark wet clothing.

A dark, foggy urban canyon with towering cracked walls and a small solitary figure standing on a concrete bridge, casting an eerie, foreboding atmosphere.

Snowy countryside landscape at sunset with warm golden light, two wooden houses with snow-covered roofs, tall snow-covered tree, and a path of footprints leading to the houses.

Futuristic alien-like humanoid figure with sleek black helmet and mechanical arms, dressed in black and gold, standing in a desert with sand dunes background.

Distorted humanoid figure made of shattered glass and metal fragments standing on a rainy city street with tall buildings and blurred pedestrians in the background.

A black cat sitting on the ground outdoors, looking up with attentive eyes, surrounded by soft-focus soil and dry leaves

Close-up of a glowing DNA double helix textured with small pinkish glowing beads, set against an ethereal blue and pink bokeh background with soft lighting effects.

Prompts Recomendados

Extreme close-up photo of a tiger eye. large title "Chroma1" overlayed in the center of the image

Prompts Negativos Recomendados

low quality, ugly, unfinished, out of focus

Parámetros Recomendados

samplers

Heun

steps

cfg

Consejos

Model is still training; community feedback is welcome.

Quantization options include FP8 Scaled Quant for ComfyUI and GGUF Quantized with ComfyUI-GGUF custom node.

Visit the Hugging Face repo for the latest model versions and training code.

Support ongoing training efforts via Ko-fi or ETH donations.

Patrocinadores del Creador

Hey everyone!

Chroma is a 8.9B parameter model based on FLUX.1-schnell (technical report coming soon!). It’s fully Apache 2.0 licensed, ensuring that anyone can use, modify, and build on top of it—no corporate gatekeeping.

The model is still training right now, and I’d love to hear your thoughts! Your input and feedback are really appreciated.

What Chroma Aims to Do

Training on a 5M dataset, curated from 20M samples including anime, furry, artistic stuff, and photos.
Fully uncensored, reintroducing missing anatomical concepts.
Built as a reliable open-source option for those who need it.

See the Progress

Hugging Face Repo: https://huggingface.co/lodestones/Chroma (go to this repo for the latest model!)
Hugging Face Debug Repo: https://huggingface.co/lodestones/chroma-debug-development-only
Live AIM Training Logs: https://training.lodestone-rock.com/
ComfyUI Inference node [WIP]: https://github.com/lodestone-rock/flux-mod
ComfyUI workflow: https://huggingface.co/lodestones/Chroma/resolve/main/simple_workflow.json
Training code!: https://github.com/lodestone-rock/flow

Quantization options

Alternative option: FP8 Scaled Quant (Format used by ComfyUI with possible inference speed increase)
Alternative option: GGUF Quantized (You will need to install ComfyUI-GGUF custom node)