B R I A 2.3
briaaiIntroduction
BRIA 2.3 is a text-to-image model developed by BRIA AI, designed for commercial applications with a focus on legal compliance and ethical content generation. It offers full legal liability coverage and is trained on a licensed dataset, ensuring high-quality, compliant imagery for business purposes.
Architecture
BRIA 2.3 is a latent diffusion text-to-image model. It utilizes a patented attribution engine for compensating data partners and is enterprise-ready with customizable technology options, allowing access to source code and weights for tailored business solutions.
Training
The model is trained exclusively on a professional-grade, licensed dataset, which excludes copyrighted materials, ensuring ethical and legally compliant outcomes.
Guide: Running Locally
-
Install Required Packages:
pip install diffusers
-
Load the Model:
from diffusers import DiffusionPipeline import torch pipe = DiffusionPipeline.from_pretrained("briaai/BRIA-2.3", torch_dtype=torch.float16, use_safetensors=True) pipe.force_zeros_for_empty_prompt = False pipe.to("cuda")
-
Generate an Image:
prompt = "A portrait of a Beautiful and playful ethereal singer, golden designs, highly detailed, blurry background" negative_prompt = "Logo,Watermark,Text,Ugly,Morbid,Extra fingers,Poorly drawn hands,Mutation,Blurry,Extra limbs,Gross proportions,Missing arms,Mutated hands,Long neck,Duplicate,Mutilated,Mutilated hands,Poorly drawn face,Deformed,Bad anatomy,Cloned face,Malformed limbs,Missing legs,Too many fingers" images = pipe(prompt=prompt, negative_prompt=negative_prompt, height=1024, width=1024).images[0]
Tips for Inference:
- Set
pipe.force_zeros_for_empty_prompt = False
. - Use negative prompts for better results.
- Use 30-50 steps for BRIA 2.3; use 8 steps for the Fast model.
- Use a
guidance_scale
of 5.0 or 7.5 for BRIA 2.3; 1.0 for the Fast model.
Cloud GPU recommendation: Consider using cloud GPU services like AWS, Google Cloud, or Azure for optimal performance.
License
BRIA 2.3 operates under commercial licensing terms, requiring a purchase for access and use. The licensing covers legal liabilities and ensures royalty management. For more details, refer to BRIA AI's terms and conditions.