Image Generation Model

Stable Image Ultra

Stable Image Ultra is a text-to-image generation model from Stability AI, supporting multiple aspect ratios and output formats.

Start Building with Stable Image Ultra View All Models

Publisher

Stability

TypeImage

Context Window77 tokens

Released2024

Price$0.08–$0.14/image

Try Stable Image Ultra →

About Stable Image Ultra

High-resolution text-to-image generation from Stability

Stable Image Ultra is a text-to-image generation model developed by Stability AI and released in 2024. It accepts text prompts along with negative prompts to guide composition, supports selectable aspect ratios, and outputs images in configurable file formats. The model operates with a 77-token context window for prompt input, which is consistent with CLIP-based text encoding used in diffusion model pipelines.

Stable Image Ultra is designed for use cases that require detailed prompt control and flexible output configuration, including creative design, concept visualization, and content production workflows. Its support for seed values allows users to reproduce specific outputs, making it useful in iterative creative processes where consistency across generations matters. The model is available as a first-party offering through Stability AI and is accessible via MindStudio without requiring separate API key management.

Capabilities

What Stable Image Ultra supports

Text-to-Image Generation

Generates images from natural language text prompts using a 77-token context window for prompt encoding.

Negative Prompting

Accepts a separate negative prompt input to suppress unwanted elements or styles from the generated image.

Aspect Ratio Selection

Allows users to choose from multiple aspect ratio presets to control the dimensions of the output image.

Output Format Control

Supports selectable output file formats, giving users control over how generated images are delivered.

Seed-Based Reproducibility

Accepts a seed value as input so that specific image outputs can be reproduced consistently across generation runs.

Ready to build with Stable Image Ultra?

Get Started Free

FAQ

Common questions about Stable Image Ultra

What is the context window size for Stable Image Ultra?

Stable Image Ultra has a context window of 77 tokens, which is the maximum prompt length the model processes for text-to-image generation.

What input types does Stable Image Ultra accept?

The model accepts a text prompt, a negative prompt, an aspect ratio selection, a seed value, and an output format selection.

Can I reproduce the same image output multiple times?

Yes. The model supports a seed input, which allows you to specify a fixed seed value and reproduce the same image output given the same prompt and settings.

What output formats does Stable Image Ultra support?

The model includes an output format selector, though the specific format options available are configured within the MindStudio interface. Common formats for Stability AI image models include PNG, JPEG, and WebP.

Is there a published price for using Stable Image Ultra on MindStudio?

Pricing information is not listed in the available metadata. You should check MindStudio's pricing page or Stability AI's official documentation for current cost details.

Resources