...
https://huggingface.co/lodestones/Chroma1-Base
https://huggingface.co/lodestones/Chroma1-HD
https://huggingface.co/lodestones/Chroma1-Radiance
https://civitai.com/models/1330309/chroma
https://civitai.com/models/1956762/chroma1-hd-gguf-and-fp8https://huggingface.co/lodestones/Chroma1-HD
| Code Block |
|---|
import torch
from diffusers import ChromaPipeline
pipe = ChromaPipeline.from_pretrained("lodestones/Chroma1-HD", torch_dtype=torch.bfloat16)
pipe.enable_model_cpu_offload()
prompt = [
"A high-fashion close-up portrait of a blonde woman in clear sunglasses. The image uses a bold teal and red color split for dramatic lighting. The background is a simple teal-green. The photo is sharp and well-composed, and is designed for viewing with anaglyph 3D glasses for optimal effect. It looks professionally done."
]
negative_prompt = ["low quality, ugly, unfinished, out of focus, deformed, disfigure, blurry, smudged, restricted palette, flat colors"]
image = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
generator=torch.Generator("cpu").manual_seed(433),
num_inference_steps=40,
guidance_scale=3.0,
num_images_per_prompt=1,
).images[0]
image.save("chroma.png") |
...
Time: 3m 51.63s | total 306.36 pipeline 230.87 callback 45.67 preview 25.74 prompt 1.51 te 1.51 decode 0.75 gc 0.30 | GPU 27140 MB 22% | RAM 38.09 GB 31%
Prompt: Bright colors, close-up photograph of a single tiger eye, direct frontal view. The iris is very detailed, and the pupil resembles a dark void. The word "Chroma" is across the upper 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. "9B" is handwritten over the image in the top left corner.
Negative: low quality, ugly, unfinished, out of focus, deformed, disfigure, blurry, smudged, restricted palette, flat colors
Parameters: Steps: 14| Size: 512x512| Seed: 21| CFG scale: 2.2| App: SD.Next| Version: 3735e97| Pipeline: ChromaPipeline| Operations: txt2img| Model: Chroma1-Flash
Time: 1m 18.75s | total 100.56 pipeline 78.05 callback 13.95 preview 4.56 prompt 1.50 te 1.50 decode 0.69 gc 0.29 | GPU 27176 MB 22% | RAM 43.73 GB 36%
| seed | chroma-unlocked-v48 | chroma-unlocked-v48-detail-calibrated | Chroma.1 base (512)Chroma.1 HD (1024) | chroma-unlocked-v50-annealed | Chroma1-HD | Chroma1-Flash |
|---|---|---|---|---|---|---|
| 433 | ||||||
| 21 | ||||||
| 38 | ||||||
| 42 | ||||||
| 2025 | ||||||
| 20251225 | ||||||
| 2925654389 |
Chroma1-Flash and CFG
Prompt: Bright colors, close-up photograph of a single tiger eye, direct frontal view. The iris is very detailed, and the pupil resembles a dark void. The word "Chroma" is across the upper 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. "9B" is handwritten over the image in the top left corner.
Negative: low quality, ugly, unfinished, out of focus, deformed, disfigure, blurry, smudged, restricted palette, flat colors
Parameters: Steps: 14| Size: 512x512| Seed: 21| CFG scale: 1.0| App: SD.Next| Version: 3735e97| Pipeline: ChromaPipeline| Operations: txt2img| Model: Chroma1-Flash
Time: 41.07s | total 65.53 pipeline 40.36 callback 13.40 preview 7.79 prompt 1.49 te 1.49 decode 0.70 gc 0.29 | GPU 27176 MB 22% | RAM 42.89 GB 35%
Steps: 14| Seed: 21| CFG scale: 1.0 | Steps: 14| Seed: 21| CFG scale: 1.1 | Steps: 14| Seed: 21| CFG scale: 1.2 | Steps: 14| Seed: 21| CFG scale: 1.3 | Steps: 14| Seed: 21| CFG scale: 1.4 |
Steps: 14| Seed: 21| CFG scale: 1.5 | Steps: 14| Seed: 21| CFG scale: 1.6 | Steps: 14| Seed: 21| CFG scale: 1.7 | Steps: 14| Seed: 21| CFG scale: 1.8 | Steps: 14| Seed: 21| CFG scale: 1.9 |
Steps: 14| Seed: 21| CFG scale: 2.0 | Steps: 14| Seed: 21| CFG scale: 2.1 | Steps: 14| Seed: 21| CFG scale: 2.2 | Steps: 14| Seed: 21| CFG scale: 2.3 | Steps: 14| Seed: 21| CFG scale: 2.4 |
Chroma,1 base and CFG
Steps: 40| Seed: 21| CFG scale: 1.0 | Steps: 40| Seed: 21| CFG scale: 1.1 | Steps: 40| Seed: 21| CFG scale: 1.2 | Steps: 40| Seed: 21| CFG scale: 1.3 | Steps: 40| Seed: 21| CFG scale: 1.4 |
Steps: 40| Seed: 21| CFG scale: 1.5 | Steps: 40| Seed: 21| CFG scale: 1.6 | Steps: 40| Seed: 21| CFG scale: 1.7 | Steps: 40| Seed: 21| CFG scale: 1.8 | Steps: 40| Seed: 21| CFG scale: 1.9 |
Steps: 40| Seed: 21| CFG scale: 2.0 | Steps: 40| Seed: 21| CFG scale: 2.2 | Steps: 40| Seed: 21| CFG scale: 2.4 | Steps: 40| Seed: 21| CFG scale: 2.6 | Steps: 40| Seed: 21| CFG scale: 2.8 |
Steps: 40| Seed: 21| CFG scale: 3.0 | Steps: 40| Seed: 21| CFG scale: 3.2 | Steps: 40| Seed: 21| CFG scale: 3.4 | Steps: 40| Seed: 21| CFG scale: 3.6 | Steps: 40| Seed: 21| CFG scale: 3.8 |
Face and hand for Chroma1-Base
Prompt: Create a close-up photograph of a woman's face and hand, with her hand raised to her chin. She is wearing a white blazer and has a gold ring on her finger. Her nails are neatly manicured and her hair is pulled back into a low bun. She is smiling and has a radiant expression on her face. The background is a plain light gray color. The overall mood of the photo is elegant and sophisticated. The photo should have a soft, natural light and a slight warmth to it. The woman's hair is dark brown and pulled back into a low bun, with a few loose strands framing her face.
Negative: low quality, ugly, unfinished, out of focus, deformed, disfigure, blurry, smudged, restricted palette, flat colors
Parameters: Steps: 40| Size: 512x512| Seed: 1620085323| CFG scale: 2.6| App: SD.Next| Version: 3735e97| Pipeline: ChromaPipeline| Operations: txt2img| Model: Chroma1-Base
Time: 3m 35.81s | total 276.87 pipeline 215.10 callback 40.75 preview 20.02 decode 0.70 gc 0.30 | GPU 27176 MB 22% | RAM 47.82 GB 39%
| cfg 2.4 | ||||
| cfg 2.6 | ||||
| 2.75* 50 | ||||
| cfg 3.0 |