canonical: https://jentic.com/apis/stability.ai/stability

# Stability AI API

The Stability AI REST API provides endpoints for image generation, editing, upscaling, 3D model generation, and account management. The API supports both v1 (engine-based) and v2beta (stable-image) endpoint patterns. The API exposes 27 endpoints secured with bearer authentication.

## For AI agents

Programmatically generate image using stable image ultra, generate image using stable image core. Covers 27 operations with bearer authentication.

## Scope

Does not handle payments, communications, or developer tools - use for crm only.

## Capabilities

- Generate image using Stable Image Ultra
- Upscale image using Fast Upscaler
- Start Creative Upscale
- Fetch Creative Upscale result
- Monitor Stability AI API operational status and events

## Use cases

### CRM Operations

Use the Stability AI API to perform crm operations programmatically. The API provides 27 endpoints covering core functionality including generate image using stable image ultra, generate image using stable image core, generate image using stable diffusion 3.

Example prompt: Call POST /v2beta/stable-image/generate/ultra to generate image using stable image ultra

### Automated Generate Management

Automate generate operations by combining multiple Stability AI API endpoints. Agents can generate image using stable image core and then generate image using stable diffusion 3 in a single workflow.

Example prompt: Call POST /v2beta/stable-image/generate/core to generate image using stable image core, then verify the result

### AI Agent Integration via Jentic

AI agents discover and call Stability AI API endpoints through Jentic without managing credentials directly. An agent searches for the required operation by intent, receives the matching endpoint schema, and executes the call with Jentic-managed authentication. This eliminates the need to read API documentation or handle bearer tokens manually.

Example prompt: Search Jentic for 'generate image using stable image ultra', load the operation schema, and execute with Jentic-managed credentials

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | /v2beta/stable-image/generate/ultra | Generate image using Stable Image Ultra |
| POST | /v2beta/stable-image/generate/core | Generate image using Stable Image Core |
| POST | /v2beta/stable-image/generate/sd3 | Generate image using Stable Diffusion 3 |
| POST | /v2beta/stable-image/upscale/fast | Upscale image using Fast Upscaler |
| POST | /v2beta/stable-image/upscale/conservative | Upscale image using Conservative Upscaler |
| POST | /v2beta/stable-image/upscale/creative | Start Creative Upscale |
| GET | /v2beta/stable-image/upscale/creative/result/{id} | Fetch Creative Upscale result |
| POST | /v2beta/stable-image/edit/inpaint | Inpaint an image |

## Key resources

- **Generate** — Image generation endpoints (text-to-image and image-to-image)
- **Upscale** — Image upscaling endpoints
- **Edit** — Image editing endpoints (inpaint, outpaint, erase, search-and-replace, etc.)
- **Control** — Image generation with structural control (sketch, structure, style)
- **3D** — 3D model generation endpoints

## Why Jentic

- **Setup:** Wiring the Stability AI API by hand means learning its bearer auth against api.stability.ai and polling async results for creative upscales yourself. Through Jentic you install once, import the Stability AI API from the API Directory, store the bearer token once, and your agent calls it.
- **Permission scoping:** You can limit the agent to the operations it needs, such as generating a core image or running a fast upscale, so it produces the image types you allow and never runs inpaint edits or creative upscales unless you add those operations.
- **Credential handling:** Your Stability AI bearer token is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'generate an image with Stable Image Core' or 'upscale this image', and Jentic returns the matching Stability AI operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Hubspot** — Alternative crm API
- **Salesforce** — Alternative crm API
- **Pipedrive** — Complementary crm API
- **Zoho** — Complementary crm API

## FAQ

### What authentication does the Stability AI API use?

The Stability AI API uses a Bearer token in the Authorization header. Through Jentic, these credentials are stored encrypted in your Jentic One instance and injected at execution time, so raw secrets never enter the agent context.

### Can I generate image using stable image ultra with the Stability AI API?

Yes. Use the POST /v2beta/stable-image/generate/ultra endpoint. The API returns structured JSON responses that agents can parse and act on directly.

### What are the rate limits for the Stability AI API?

Rate limits are not specified in the OpenAPI spec. Check the vendor documentation for current limits. Through Jentic, rate limiting is handled automatically with retry logic built into the execution layer.

### How do I generate image using stable image ultra through Jentic?

Install the Jentic SDK with pip install jentic, authenticate through Jentic One, the self-hosted execution layer, then search for 'generate image using stable image ultra'. Jentic returns the matching Stability AI API operation with its input schema. Load the schema and execute the call - credentials are injected automatically.

### How many endpoints does the Stability AI API have?

The Stability AI API exposes 27 endpoints covering generate, upscale, edit operations.

### Can I limit what my agent is allowed to do with the Stability AI API?

Yes. Because you self-host Jentic One, your own rules decide which Stability AI operations and credentials the agent may use. You can scope the agent to just the calls it needs, such as generating a Stable Image Core image or running a fast upscale, so it produces only the image types you allow. It never runs inpaint edits or creative upscales unless you explicitly add those operations.