models/Chroma - v.41-low-step-rl

Chroma - v.41-low-step-rl

|
8/16/2025
|
3:30:46 AM
| Discussion|
0
Close-up of a weathered scarecrow's face with vivid green eyes, rough burlap texture, strands of dry straw, and a dark worn hat with frayed edges.
Detailed oil painting of Vasilisa the Beautiful in traditional Slavic dress standing in a misty forest with Baba Yaga looming behind her, near a witch hut on chicken legs, featuring expressionist brushstrokes and a muted color palette.
Close-up of a young woman's face focusing on her luminous blue eye reflecting a burning city, featuring an ethereal elven ear and dark moody tones.
A young woman with a red backpack stands in a golden field under a bright blue sky, gazing at a colossal metallic ring structure dominating the horizon.
A cyberpunk girl with white hair featuring red streaks, wearing an exoshell mask and robotic ears, squatting in a high-tech laboratory surrounded by dark cables, holding a glowing holographic blue butterfly.
Abstract expressionist painting of a girl with blue hair and cat ears leaning on a detailed motorbike, painted with textured brushstrokes in muted tones.
A female figure in tattered clothing walks across a desert towards a towering, weathered war machine robot walker, with a futuristic city skyline in the distance under a bright blue sky.
A woman in a sleek black spacesuit sits expressionless inside a dimly lit futuristic cockpit, gazing out at a cyberpunk cityscape in black and white manga style.
Portrait of a young woman wearing a sleek metallic astronaut helmet with a transparent visor and fur-trimmed jacket, set against a snowy mountainous background.
Watercolor painting of a long-legged girl sitting on an oversized military helmet holding a leash attached to a small chunky robot, with gritty brushstrokes and muted colors.
Fantastical towering waterfront dwelling with patchwork architecture featuring vibrant red sections and blue roofs, perched on wooden pier above calm reflective water.
A young girl with pale skin and white bob hairstyle, wearing an oversized beige sweater and white sneakers, sitting on the shoulder of a large, industrial, slightly worn humanoid robot under a clear blue sky.

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 LOW STEPS RL TRAINING" is across the lower portion of the image in large white stylized letters, with brush strokes. 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

DPM++ 2M

steps

14

cfg

1.3

Tips

Model training is ongoing, and user feedback is encouraged.

Use the latest model from the Hugging Face repo https://huggingface.co/lodestones/Chroma

Alternative quantization options are available including FP8 Scaled Quant and GGUF Quantized formats.

You can follow live training progress via https://training.lodestone-rock.com/

ComfyUI inference and workflow resources are provided to facilitate integration.

Creator Sponsors

Fictional.ai Support

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

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
Cyberpunk Anime Style - SDXL v2.0 light version
Next
Hyperdetailed Illustration - SDXL

Model Details

Model type

Checkpoint

Base model

Other

Model version

v.41-low-step-rl

Model hash

f6fa0d119e

Creator

Discussion

Please log in to leave a comment.