models/Chroma - v.35-detail-calibrated

Chroma - v.35-detail-calibrated

|
7/3/2025
|
5:58:52 AM
| Discussion|
0
Dark shadowy figure with a horned skull head, screaming with black paint dripping from its mouth and eyes, depicted in a gothic horror style.
Dark-skinned cyberpunk demon with fiery orange and black scales, glowing blue cybernetic eyes, wearing a tattered white hooded jacket with an orange scarf, holding a large glowing axe with a fiery blade.
Grayscale surreal painting of a person holding a cracked mask resembling a broken face, expressing emotional suffering against a dark background.
Dark shadowed figure surrounded by walls full of black holes with various watchful eyes inside, evoking fear and surreal horror atmosphere.
Silhouette of a wolf sitting on grassy hills illuminated by a glowing sunset, with a full moon and starry night sky above.
A digital painting of a snow-covered disgruntled tiger with orange and white fur and black stripes, snow on its nose and whiskers, set against lush green and colorful spring foliage with signs of autumn and summer seasons.
Close-up digital painting of a Border Collie with dark brown and black fur, white face, and bright blue eyes, showing a tilted head with a grin on a dark brown background.
Snowy medieval city at night with warm glowing street lamps, a cloaked figure standing on a rooftop, and a grand castle illuminated in the background.
A grotesque alien-like creature with a glossy translucent dome encasing its head, featuring a prominent curved beak, rough textured red-brown decaying skin, and muscular sinewy upper body with dark bead-like protrusions.
Anime-style illustration of a person standing on a rocky cliff overlooking a vast landscape with detailed clouds and a moody atmosphere during sunset.
A glowing, animated ball with expressive eyes sits on a fiery, molten surface with dramatic lighting and cinematic shadows.
A mysterious dark cat with glowing orange eyes holding a beer bottle in a cinematic, detailed, and sharply composed painting style.

Recommended Prompts

Extreme close-up photograph of a single tiger eye, direct frontal view. The iris is very detailed and the pupil resembling a dark void. The word "Chroma V.35 now with less steps" is across the lower portion of the image in large white stylized letters, with brush strokes resembling those made with Japanese calligraphy. Each strand of the thick fur is highly detailed and distinguishable. Natural lighting to capture authentic eye shine and depth.

Recommended Negative Prompts

low quality, ugly, unfinished, out of focus

Recommended Parameters

samplers

Euler

steps

16

cfg

4.5

Tips

Chroma model is still under active training; user feedback is welcome.

Alternative quantization options available: FP8 Scaled Quant (better inference speed with ComfyUI) and GGUF Quantized (requires ComfyUI-GGUF custom node).

Model and related resources including training code, inference workflows, and logs are accessible via provided Hugging Face and GitHub links.

Creator Sponsors

Fictional.ai advertisement

Special thanks to Fictional.ai for support in advancing open-source AI.

Support ongoing training costs with donations: https://ko-fi.com/lodestonerock/goal?g=1

ETH donations address: 0x679C0C419E949d8f3515a255cE675A1c4D92A3d7

Join the community on Discord: discord.gg/SQVcWVbqKx

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

Quantization options

Special Thanks

Shoutout to Fictional.ai for the awesome support — seriously appreciate you helping push open-source AI forward.

Support Open-Source AI

The current pretraining run has already used 5000+ H100 hours, and keeping this going long-term is expensive.

If you believe in accessible, community-driven AI, any support would be greatly appreciated.

👉 [https://ko-fi.com/lodestonerock/goal?g=1] — Every bit helps!

ETH: 0x679C0C419E949d8f3515a255cE675A1c4D92A3d7

my discord: discord.gg/SQVcWVbqKx

Previous
Hassaku XL (Illustrious) - v1.3 - Style A
Next
748cm | Style for Illustrious/NoobAI 0.75 - v1.0

Model Details

Model type

Checkpoint

Base model

Other

Model version

v.35-detail-calibrated

Model hash

e01d54637f

Creator

Discussion

Please log in to leave a comment.