pkg.gl

Categorygithub.com/instill-ai/pipeline-backendpkgcomponentaistabilityaiv0

package

0.40.0-beta

Repository: https://github.com/instill-ai/pipeline-backend.git

Documentation: pkg.go.dev

# README

title: "Stability AI" lang: "en-US" draft: false description: "Learn about how to set up a VDP Stability AI component https://github.com/instill-ai/instill-core"

The Stability AI component is an AI component that allows users to connect the AI models served on the Stability AI Platform. It can carry out the following tasks:

Text to Image
Image to Image

Release Stage

Alpha

Configuration

The component definition and tasks are defined in the definition.json and tasks.json files respectively.

Setup

In order to communicate with Stability AI, the following connection details need to be provided. You may specify them directly in a pipeline recipe as key-value pairs within the component's setup block, or you can create a Connection from the Integration Settings page and reference the whole setup as setup: ${connection.<my-connection-id>}.

Field	Field ID	Type	Note
API Key	`api-key`	string	Fill in your Stability AI API key. To find your keys, visit here

Supported Tasks

Text to Image

Generate a new image from a text prompt.

Input	ID	Type	Description
Task ID (required)	`task`	string	`TASK_TEXT_TO_IMAGE`
Engine (required)	`engine`	string	Stability AI Engine (model) to be used.
Prompts (required)	`prompts`	array[string]	An array of prompts to use for generation.
Weights	`weights`	array[number]	An array of weights to use for generation.
CFG Scale	`cfg-scale`	number	How strictly the diffusion process adheres to the prompt text (higher values keep your image closer to your prompt)
Clip Guidance Preset	`clip-guidance-preset`	string	Clip guidance preset
Height	`height`	integer	The image height
Width	`width`	integer	The image width
Sampler	`sampler`	string	Which sampler to use for the diffusion process. If this value is omitted we'll automatically select an appropriate sampler for you.
Samples	`samples`	integer	Number of images to generate
Seed	`seed`	integer	Random noise seed (omit this option or use `0` for a random seed)
Steps	`steps`	integer	Number of diffusion steps to run.
Style Preset	`style-preset`	string	Pass in a style preset to guide the image model towards a particular style. This list of style presets is subject to change.

Output	ID	Type	Description
Images	`images`	array[string]	Generated images
Seeds	`seeds`	array[number]	Seeds of generated images

Image to Image

Modify an image based on a text prompt.

Input	ID	Type	Description
Task ID (required)	`task`	string	`TASK_IMAGE_TO_IMAGE`
Engine (required)	`engine`	string	Stability AI Engine (model) to be used.
Prompts (required)	`prompts`	array[string]	An array of prompts to use for generation.
Init Image	`init-image`	string	Image used to initialize the diffusion process, in lieu of random noise.
Weights	`weights`	array[number]	An array of weights to use for generation. If unspecified, the model will automatically assign a default weight of 1.0 to each prompt.
Clip Guidance Preset	`clip-guidance-preset`	string	Clip guidance preset
Image Strength	`image-strength`	number	How much influence the `init_image` has on the diffusion process. Values close to `1` will yield images very similar to the `init_image` while values close to `0` will yield images wildly different than the `init_image`. The behavior of this is meant to mirror DreamStudio's "Image Strength" slider. This parameter is just an alternate way to set `step_schedule_start`, which is done via the calculation `1 - image_strength`. For example, passing in an Image Strength of 35% (`0.35`) would result in a `step_schedule_start` of `0.65`.
Cfg Scale	`cfg-scale`	number	How strictly the diffusion process adheres to the prompt text (higher values keep your image closer to your prompt)
Init Image Mode	`init-image-mode`	string	Whether to use `image_strength` or `step_schedule_*` to control how much influence the `init_image` has on the result.
Sampler	`sampler`	string	Which sampler to use for the diffusion process. If this value is omitted we'll automatically select an appropriate sampler for you.
Samples	`samples`	integer	Number of images to generate
Seed	`seed`	integer	Random noise seed (omit this option or use `0` for a random seed)
Step Schedule Start	`step-schedule-start`	number	Skips a proportion of the start of the diffusion steps, allowing the init_image to influence the final generated image. Lower values will result in more influence from the init_image, while higher values will result in more influence from the diffusion steps. (e.g. a value of `0` would simply return you the init_image, where a value of `1` would return you a completely different image.)
Step Schedule End	`step-schedule-end`	number	Skips a proportion of the end of the diffusion steps, allowing the init_image to influence the final generated image. Lower values will result in more influence from the init_image, while higher values will result in more influence from the diffusion steps.
Steps	`steps`	integer	Number of diffusion steps to run.
Style Preset	`style-preset`	string	Pass in a style preset to guide the image model towards a particular style. This list of style presets is subject to change.