Skip to main content

​
About the Model

Stability AI is a generative AI company known for creating and advancing the Stable Diffusion family of image generation models. The organization focuses on open-source and broadly accessible AI tools that enable high-quality text-to-image generation and creative workflows for developers, researchers, and creators worldwide.

​
Model Quickstart

This section helps you quickly get started with the stabilityai/stable-diffusion-3.5-large model on the Qubrid AI inferencing platform. To use this model, you need:
  • A valid Qubrid API key
  • Access to the Qubrid inference API
  • Basic knowledge of making API requests in your preferred language
Once authenticated with your API key, you can send inference requests to the stabilityai/stable-diffusion-3.5-large model and receive responses based on your input prompts. Below are example placeholders showing how the model can be accessed using different programming environments.
You can choose the one that best fits your workflow. 
import requests
import json

url = "https://platform.qubrid.com/api/v1/qubridai/image/generation"
headers = {
"Authorization": "Bearer <QUBRID_API_KEY>",
"Content-Type": "application/json"
}

data = {
"model": "stabilityai/stable-diffusion-3.5-large",
"positive_prompt": "cinematic shot of a lone astronaut standing on a desolate alien planet, glowing orange sunset sky, dust storms swirling, dramatic lighting, ultra-wide lens composition, movie still aesthetic, realistic space suit details, volumetric atmosphere, 8k sci-fi film scene",
"width": 1024,
"height": 1024,
"steps": 30,
"cfg": 7.5,
"seed": 42
}

response = requests.post(url, headers=headers, json=data)

if response.status_code == 200:
    with open("generated_image.png", "wb") as f:
        f.write(response.content)
    print("Image saved to generated_image.png")
else:
    print(f"Error {response.status_code}: {response.text}")

​
API Generated Response

​
Example Generated Image

Prompt: ethereal elf queen with silver hair and glowing emerald eyes, standing in enchanted forest, magical aura around her, cinematic fantasy lighting, high-detail armor and jewelry, digital art, concept art style, artstation trending

​
Model Overview

Stable Diffusion 3.5 Large is a text-to-image generation model designed for high-quality image synthesis with strong prompt adherence. With 8.1 billion parameters, it is the most powerful base model in the Stable Diffusion family and is intended for professional image generation at 1-megapixel resolution. The model focuses on producing visually rich images across a wide range of styles while maintaining accurate interpretation of complex text prompts. The model is built as a Multimodal Diffusion Transformer (MMDiT) and improves image quality, typography handling, complex prompt understanding, and overall resource efficiency compared to earlier Stable Diffusion versions.

​
Model at a Glance

FeatureDetails
Model Namestabilityai/stable-diffusion-3.5-large
ProviderStability AI
Model Size8.1B params
ArchitectureLatent Diffusion Model (LDM) with a UNet backbone and dual CLIP text encoders (OpenCLIP ViT-G and CLIP ViT-L)
Default Resolution1024 × 1024
Training DataLarge-scale image–text datasets (including LAION, aesthetic filtering)
Output TypeImages generated from text prompts

​
When to use?

You should consider using Stable Diffusion 3.5 Large if:
  • You need high-quality text-to-image generation at 1-megapixel resolution
  • Your application requires strong prompt adherence and complex prompt understanding
  • You want support for multiple visual styles such as photography, illustration, 3D, painting, or line art
  • You need reliable text rendering and typography within generated images
This model is intended for creative, design, and research-focused image generation workflows rather than factual or real-world verification tasks.

​
Inference Parameters

Parameter NameTypeDefaultDescription
Widthnumber1024Image width in pixels.
Heightnumber1024Image height in pixels.
Inference Stepsnumber30Number of denoising steps.
Guidance Scalenumber7.5How closely to follow the prompt.
Seednumber50Random seed for reproducibility.
Negative Promptstring—Specifies what to exclude from the image.

​
Key Features

  • High-Quality Image Generation: 8.1B parameter model designed for professional-grade image generation at 1MP resolution.
  • Strong Prompt Adherence: Improved understanding of complex prompts, typography, and detailed instructions.
  • Versatile Visual Styles: Supports a wide range of styles including photography, illustration, 3D, painting, and line art.
  • Diverse Image Outputs: Generates images representing varied appearances and features without extensive prompt engineering.
  • Efficient Diffusion Architecture: Built on an MMDiT-based latent diffusion framework for better quality and resource efficiency.

​
Summary

Stable Diffusion 3.5 Large is the most powerful base model in the Stable Diffusion family, built for high-quality text-to-image generation.
  • It focuses on strong prompt adherence, improved typography, and detailed visual understanding.
  • The model supports diverse visual styles and produces consistent results at 1-megapixel resolution.
  • It is suitable for professional creative workflows such as design, illustration, and artistic generation.
  • The model is not intended for factual accuracy or real-world verification tasks.