A LLaVA vision-language model running on Cloudflare Workers AI, on the same platform as the API, so no image is sent to a third-party provider.

How do I send an image?

Either a public image URL with a GET request, or the raw bytes in a POST. JPEG, PNG and WebP up to 2.5 MB are accepted.

Can I ask a specific question?

Yes. Pass a prompt such as 'what colour is the car' or 'what does the sign say' and the model answers about that image rather than giving a generic description.

How accurate is text reading?

Prominent, clearly printed text is read well. Small print, handwriting and dense documents are better served by a dedicated OCR tool.

What does a call cost?

25 tokens, reflecting the GPU inference behind each description. The free plan covers 20 images a month.

AIMAINTAINED25 tokens / call

Image to Text API

Describe any image, answer questions about it or read its text with an on-platform vision model

Describe an image, answer a question about it or read the prominent text in it with a vision-language model running on the platform. Send an image URL or POST the bytes, JPEG, PNG or WebP up to 2.5 MB. No third-party keys and no data leaves the platform.

Every call needs an X-Api-Key header. A free key includes 500 tokens a month, no card required. 25 tokens per call, so the free plan covers about 20 calls a month.

Start free — 500 tokens / mo →See all plans

No credit card requiredAccess to 125+ APIsCancel anytime

response.json

{
  "ok": true,
  "prompt": "Describe this image in detail.",
  "text": "The image features a small, cute cat figurine with a brown and white color scheme. The cat is standing on its hind legs, giving the impression of a real cat. The figurine is positioned in the center of the image, and its unique design makes it an appealing and charming decoration."
}

Production Ready: Battle-tested infrastructure with a 99.9% uptime SLA. Rate limits are generous, responses are consistent.
Simple Integration: Standard REST + JSON. Works with any HTTP client. Copy-ready code snippets in 4 languages.
One Key, 125+ APIs: Your DataAndAPIs subscription unlocks every API on the platform — anime, finance, sports, and more.

Use cases for Image to Text API

Generate alt text for uploaded images automatically to meet accessibility requirements.
Let users ask questions about a photo, such as what is in it or what a sign says.
Extract prominent text from screenshots, labels and signs without a separate OCR service.
Tag and caption large image libraries for search and moderation.

Image to Text API endpoints

Every endpoint with its parameters and a real example response.

GETDescribe Image
POSTDescribe Image (POST)

GETDescribe Image

https://api-image-to-text.dataandapis.com/v1/describe

Query parameters

Parameter	Example
prompt	Describe this image in detail.	Optional
url	https://dataandapis.com/images/cats.webp	Optional

Example request

curl "https://api-image-to-text.dataandapis.com/v1/describe?prompt=Describe%20this%20image%20in%20detail.&url=https%3A%2F%2Fdataandapis.com%2Fimages%2Fcats.webp" \
  -H "X-Api-Key: YOUR_API_KEY"

Example response

{
  "ok": true,
  "prompt": "Describe this image in detail.",
  "text": "The image features a small, cute cat figurine with a brown and white color scheme. The cat is standing on its hind legs, giving the impression of a real cat. The figurine is positioned in the center of the image, and its unique design makes it an appealing and charming decoration."
}

POSTDescribe Image (POST)

https://api-image-to-text.dataandapis.com/v1/describe

Request body

{
  "url": "https://dataandapis.com/images/cats.webp",
  "prompt": "What animal is in this picture? Answer in a few words."
}

Example request

curl -X POST "https://api-image-to-text.dataandapis.com/v1/describe" \
  -H "X-Api-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "url": "https://dataandapis.com/images/cats.webp", "prompt": "What animal is in this picture? Answer in a few words." }'

Example response

{
  "ok": true,
  "prompt": "What animal is in this picture? Answer in a few words.",
  "text": "Cat"
}

Where the data comes from

Source: Cloudflare Workers AI vision-language model (LLaVA 1.5 7B). No stored dataset.
Freshness: Inference runs live on every call.

One install. Every endpoint.

No proprietary SDKs needed. Any HTTP client works. Pick your language and paste.

3languages

2endpoints

RESTarchitecture

1key for all APIs

Fully typed responses — auto-completion in your IDE
Tracks the API in lockstep — new endpoints land same day
One key works across 125+ APIs on the platform

import axios from 'axios';

const { data } = await axios.get(
  'https://api-image-to-text.dataandapis.com/v1/describe',
  { headers: { 'X-Api-Key': process.env.API_KEY } }
);

Image to Text API · X-Api-Key

Explore the endpoints

Browse request parameters and inspect live example responses.

Available Endpoints

Request Parameters

API URL

Headers

X-Api-Key

Query Parameters

prompt(Optional)

url(Optional)

Target:

Client:

import axios from 'axios';

const url = 'https://api-image-to-text.dataandapis.com/v1/describe';
const { data } = await axios.get(url, {
  headers: { 'X-Api-Key': 'YOUR_API_KEY' }
});