Chat with any image.
Upload a photo and ask anything — the AI describes it, extracts text, identifies objects, and holds a real conversation about what it sees.
Upload a photo and ask questions. The AI will describe, analyze, and explain anything in your image.
Vision, OCR, and chat
in one endpoint.
messages array to keep context. Follow-up questions understand the image.url, or a data:image/... base64 payload. Whichever fits.curl and call it./api/imgchat. GET for health, POST for the AI reply.From photo to answer
in seconds.
Free REST API.
Single endpoint.
CORS-enabled. No key. GET for health, POST for AI answers. Both accept JSON.
/api/imgchat
health check
Health check — confirms the service is up and returns basic status JSON.
/api/imgchat
chat
Send an image and a question. The AI analyzes and returns a natural-language answer. Supports OCR, object detection, text extraction, and multi-turn conversation.
| Field | Type | Description | |
|---|---|---|---|
| userPrompt | string | required | The question or instruction about the image. |
| image | string | optional | Base64 data URL — data:image/jpeg;base64,… (use this or url). |
| url | string | optional | Public image URL (use this or image). |
| messages | array | optional | Previous turns for multi-turn chat. Each item: { type:"user"|"ai", content:"…" }. |
{ "success": true, "response": "The image shows a golden retriever..." }
image (base64 data URL) or url (public link). Supported: JPEG, PNG, WEBP, GIF. Max ~10MB before encoding./api/imgchat
multi-turn
To keep context across follow-up questions, pass the full messages history on every request.
fileToBase64()
encoding
Turn a File-input value into the base64 data URL the endpoint expects.
function fileToBase64(file){
return new Promise((resolve, reject) => {
const reader = new FileReader();
reader.onload = () => resolve(reader.result);
reader.onerror = reject;
reader.readAsDataURL(file);
});
}
const file = document.querySelector('input[type=file]').files[0];
const dataUrl = await fileToBase64(file);
// "data:image/jpeg;base64,/9j/4AAQ..."
Error codes
Frequently asked
questions.
What is VisionSter?
A free AI vision chat. Upload any image and the model will describe, OCR, analyze and answer follow-up questions in plain language.
Is it really free?
Yes. Free in-browser, and free via the REST API. No sign-up, no API key, no credit card.
Which image formats are supported?
JPEG, PNG, WEBP and GIF, up to ~10MB before base64 encoding. Larger files will be rejected with a 413.
Can I use a public URL instead of uploading?
Yes. Send url instead of image and the endpoint will fetch it server-side.
How do I keep context across follow-ups?
Pass a messages array on every request, ordered from earliest to latest, with items like { type:"user", content:"…" }.
Can it read text (OCR)?
Yes. Just ask "extract all text" or "transcribe this receipt" — no separate OCR service needed.
Follow our
WhatsApp channels.
New tools, updates, and exclusive resources delivered to WhatsApp.
About · Privacy · Terms.
VisionSter is a free AI image-chat service by AHM7xMakki — part of a 16-tool free API platform. Owned by AHM7. Built and managed by Makki. About us →
Images are forwarded to the AI provider to generate the response and are not persisted beyond the request. Full policy →
For lawful personal and commercial use. Don't upload illegal or non-consensual content. Fair-use limits apply. Full terms →