metadata

language: en
license: apache-2.0
tags:
  - art
  - image-generation
  - text-to-image
  - diffusion
  - high-quality
  - AI
pipeline_tag: text-to-image
thumbnail: https://example.com/thumbnail.png

ImageGroundV0.1

Model Card for enhanceaiteam/ImageGroundV0.1

ImageGroundV0.1 is a powerful text-to-image generation model developed by EnhanceAI. It leverages advanced diffusion techniques to produce high-quality, detailed images based on textual descriptions. This model is designed to assist artists, designers, and creative professionals in generating visually compelling and accurate images.

Model Details

Version: 0.1
Base Model: EnhanceAI's proprietary image generation architecture
Training Data: The model has been trained on a diverse dataset of high-resolution, copyright-free images, ensuring a broad range of visual styles and subjects.
Pipeline: Stable Diffusion XL Pipeline

Features

High-Resolution Output: Generates images at 1024x1024 resolution.
Enhanced Prompt Understanding: Improved comprehension of complex and detailed text prompts.
Versatile Style Generation: Capable of producing images in various artistic and realistic styles.
High-Quality Images: Delivers visually stunning and detailed images.

Usage

To use ImageGroundV0.1, you can integrate it with the diffusers library. Below is an example of how to generate images:

import torch
from diffusers import (
    StableDiffusionXLPipeline,
    KDPM2AncestralDiscreteScheduler,
    AutoencoderKL
)

# Load VAE component
vae = AutoencoderKL.from_pretrained(
    "madebyollin/sdxl-vae-fp16-fix", 
    torch_dtype=torch.float16
)

# Configure the pipeline
pipe = StableDiffusionXLPipeline.from_pretrained(
    "enhanceaiteam/ImageGroundV0.1", 
    vae=vae,
    torch_dtype=torch.float16
)
pipe.scheduler = KDPM2AncestralDiscreteScheduler.from_config(pipe.scheduler.config)
pipe.to('cuda')

# Define prompts and generate image
prompt = "a serene landscape with mountains and a clear sky"
negative_prompt = ""

image = pipe(
    prompt, 
    negative_prompt=negative_prompt, 
    width=1024,
    height=1024,
    guidance_scale=7,
    num_inference_steps=50,
    clip_skip=2
).images[0]

image.save("generated_image.png")

Prompts

When creating prompts, be as descriptive as possible to achieve the desired results. Detailed descriptions help the model generate more accurate and relevant images.