canonical: https://jentic.com/apis/alttext.ai/alttext-ai

# AltText.ai API

Jentic publishes the only available OpenAPI specification for AltText.ai API, keeping it validated and agent-ready. AltText.ai is a vision-AI service that generates accessibility-grade alt text for images at scale, with support for over a hundred languages and an e-commerce-aware mode that incorporates product context. The API exposes 10 X-API-Key endpoints for creating images individually or in bulk, scraping pages for image discovery, and managing the resulting asset records. Teams use it to retrofit alt text on existing CMS libraries and to generate alt text on upload for new content.

## For AI agents

Generate accessibility-grade alt text for images, including bulk and page-scrape modes, through 10 X-API-Key endpoints. Useful for agents adding alt text to CMS libraries, e-commerce catalogues, and email assets.

## Scope

Does not host images, run general-purpose visual question answering, or perform image moderation - use only for generating accessibility-grade alt text from supplied image URLs.

## Capabilities

- Generate alt text for a single image via POST /images, with optional language, gpt_prompt, and keyword hints
- Bulk-generate alt text for many images in one request through POST /images/bulk_create
- Scrape a public web page for image URLs and generate alt text for each via POST /images/page_scrape
- Search through previously generated images by metadata using GET /images/search
- Update or delete a stored image record by asset id
- Read and update the AltText.ai account profile and language defaults
- Request alt text in multiple languages in a single call by passing a comma-separated lang parameter

## Use cases

### Retrofit Alt Text onto a CMS Library

Editorial and accessibility teams retrofit alt text onto thousands of legacy CMS images without writing each description by hand. The /images/bulk_create endpoint accepts a list of image URLs, AltText.ai generates descriptions, and the integration writes the results back into the CMS. Multi-language support lets the same job populate alt text for every locale the site supports.

Example prompt: POST a JSON body to /images/bulk_create with 200 image URLs and lang='en,es,fr', then write each returned description into the CMS field for the matching image.

### On-Upload Alt Text for E-Commerce Catalogues

E-commerce teams call POST /images on every new product upload, passing keywords and the gpt_prompt parameter to nudge the model toward product-aware descriptions. Generated alt text lands on the storefront automatically, which improves accessibility scores and image-search SEO without adding a manual review step for routine SKUs.

Example prompt: On product image upload, POST /images with the image URL, the SKU's product title as gpt_prompt, and keywords from the product tags.

### Page-Scrape Alt Text Audits

Accessibility consultants run an audit by pointing /images/page_scrape at a target URL; AltText.ai discovers every image on the page and returns generated descriptions side by side with the existing alt attributes. The team can quickly highlight images that are missing alt text or have generic placeholders.

Example prompt: POST /images/page_scrape with the audit URL, then diff the returned alt-text suggestions against the page's existing alt attributes to flag gaps.

### AI Agent Accessibility Worker via Jentic

An AI agent built on Jentic can take 'add alt text to every image in this folder' and resolve it to a /images/bulk_create call with the X-API-Key safely vaulted in your Jentic One instance. The agent handles paging, retries, and writeback to the source system through Jentic's execution layer.

Example prompt: Search Jentic for 'bulk generate alt text', load the POST /images/bulk_create schema, batch the input image URLs in groups of 200, and execute each batch.

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | /images | Generate alt text for a single image |
| GET | /images | List previously generated alt-text records |
| GET | /images/{asset_id} | Get a single alt-text record by asset id |
| PUT | /images/{asset_id} | Update an existing alt-text record |
| DELETE | /images/{asset_id} | Delete a stored alt-text record |
| GET | /images/search | Search alt-text records by metadata |
| POST | /images/bulk_create | Bulk-generate alt text for many images in one call |
| POST | /images/page_scrape | Scrape a page for images and generate alt text for each |

## Key resources

- **Images** — Generate, search, update, and delete alt-text image records
- **Account** — Read and update AltText.ai account settings and default language

## Why Jentic

- **Setup:** Wiring the AltText.ai API by hand means attaching its API key to alttext.ai/api/v1 and encoding the image create, retrieve, and bulk calls yourself. Through Jentic you install once, import the AltText.ai API from the API Directory, store the API key once, and your agent calls it.
- **Permission scoping:** The AltText.ai API puts the asset id in the URL path (/images/{asset_id}), so a rule can pin your agent to reading one image asset. You choose the operations it may call, so bulk creation, updates, or deletion is not included unless you add them.
- **Credential handling:** Your AltText.ai API key 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 alt text for an image' or 'retrieve alt text for an asset', and Jentic returns the matching AltText.ai operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **OpenAI API** — Use a general vision model (e.g., GPT-4o) to caption images with full prompt control
- **DeepAI API** — General-purpose AI APIs including image captioning and tagging
- **AltTextLab API** — AI alt-text generation with 130+ languages and an e-commerce mode
- **Cloudinary Upload API** — Image storage, transformation, and CDN delivery to host the assets being captioned

## FAQ

### Why is there no official OpenAPI spec for AltText.ai API?

AltText.ai does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call AltText.ai API via structured tooling. It is validated against the live API and kept up to date. Get started with Jentic One, the self-hosted execution layer.

### What authentication does the AltText.ai API use?

AltText.ai uses an X-API-Key header on every request. Through Jentic, the key is stored encrypted in your Jentic One instance and injected at call time, so the raw key never enters agent prompts or shared logs.

### Can I generate alt text in multiple languages with the AltText.ai API?

Yes. Pass a comma-separated list to the lang parameter (e.g., lang=en,es,fr) on POST /images, /images/bulk_create, or /images/page_scrape. Each additional language beyond the first consumes one translation credit per image, and gpt_prompt and keywords must be supplied in English.

### What are the rate limits for the AltText.ai API?

Some endpoints are rate-limited and return 429 Too Many Requests when limits are reached. The exact ceiling is tied to your AltText.ai plan; build retry-with-backoff and prefer /images/bulk_create over many single-image calls when processing a backlog.

### How do I bulk-generate alt text for a CMS library through Jentic?

Run pip install jentic and search with 'bulk generate alt text'. Load the POST /images/bulk_create schema, pass an array of source image URLs, and execute. Jentic handles auth and retries; your code only handles writeback to the CMS.

### Can the AltText.ai API discover images on a page automatically?

Yes. POST /images/page_scrape with a target URL; AltText.ai discovers the images on the page and returns generated alt text for each, which is useful for accessibility audits and migrations.

### Can I limit what my agent is allowed to do with the AltText.ai API?

Yes. Because you run Jentic One yourself, your own rules decide which AltText.ai operations and credentials the agent can use. Since the AltText.ai API puts the asset id in the URL path (/images/{asset_id}), a rule can pin the agent to reading a single image record while excluding write operations. Bulk generation via POST /images/bulk_create, updates via PUT /images/{asset_id}, and deletion via DELETE /images/{asset_id} are only available to the agent if you explicitly grant them.