# Moondream 3.1 on Workers AI: fast vision at the edge

> Cloudflare brings the Moondream 3.1 vision model to Workers AI. Caption images, ask questions, get object coordinates and bounding boxes — in under a second.

Author: [Flavio Copes](https://flaviocopes.com/about/) | Published: 2026-07-09 | Topics: [Cloudflare](https://flaviocopes.com/tags/cloudflare/) | Canonical: https://flaviocopes.com/moondream-workers-ai/

Workers AI just got eyes.

Cloudflare [partnered with Moondream](https://developers.cloudflare.com/changelog/post/2026-07-08-moondream31-workers-ai/) to bring their latest vision language model to Workers AI. You can now send an image to a Worker and ask questions about it, caption it, or get the coordinates of objects inside it — with responses coming back in well under a second.

The model is `@cf/moondream/moondream3.1-9B-A2B`.

## What is Moondream 3.1?

Moondream is a small, fast vision language model. Version 3.1 uses a **mixture-of-experts** architecture: 9B total parameters, but only 2B active per request.

That's the trick behind the speed. You get visual reasoning close to much bigger models, with the inference cost and latency of a small one. It also has a 32K token context window, so you can ask detailed, structured questions.

## The four skills

The model does four things, selected with a `task` parameter:

- **query** — ask open-ended questions about an image
- **caption** — generate a short, normal, or long description
- **point** — get the coordinates of objects matching a phrase
- **detect** — get bounding boxes for objects matching a phrase

The last two are what make this interesting. Most vision models describe images. Moondream can tell you *where* things are, which is what you need for robotics, UI automation, and image tooling.

And it's fast: Cloudflare reports first tokens streaming back in roughly 20–30 ms, with point and detect completing in the tens-to-low-hundreds of milliseconds range for a simple image, and full captions and queries staying under a second.

## Using it from a Worker

Same pattern as every Workers AI model: the `AI` binding and `env.AI.run()`. Here's the caption task:

```js
export default {
  async fetch(request, env) {
    const res = await fetch('https://cataas.com/cat')
    const blob = await res.arrayBuffer()

    const response = await env.AI.run('@cf/moondream/moondream3.1-9B-A2B', {
      image: [...new Uint8Array(blob)],
      prompt: 'Generate a caption for this image',
      max_tokens: 512,
    })

    return Response.json(response)
  },
}
```

The `image` field also accepts a public HTTPS URL or a base64 data URI, so you don't have to fetch and convert the bytes yourself.

For object detection, switch the task:

```js
const response = await env.AI.run('@cf/moondream/moondream3.1-9B-A2B', {
  task: 'detect',
  image: 'https://example.com/team-photo.jpg',
  prompt: 'face',
})
```

You get back bounding box coordinates for every match.

## What this unlocks

The combination of fast + cheap + edge changes which vision features are practical:

- **moderating user uploads** before they're stored — a caption or query call as a synchronous step in the upload path
- **alt text generation** for images, at scale
- **visual agents** that look at screenshots and decide what to click
- **live feeds** — camera frames, robotics — where a multi-second round trip to a big model is a dealbreaker

I generate alt text for images on this site with a vision model in a batch script. At these latencies you don't need the batch — you can do it at request time.

## Pricing

Moondream 3.1 costs **$0.30 per million input tokens** and **$1.00 per million output tokens** on Workers AI.

Vision inputs tokenize images, so input tokens dominate — but a detect call that returns a few coordinates produces almost no output tokens. Point-and-detect workloads are close to free at small scale, and the Workers AI free daily allocation applies here too.

## Trying it

Check it's in the catalog and you're on a current wrangler:

```bash
npx wrangler ai models | grep moondream
```

Then add the `AI` binding to your `wrangler.jsonc` and call it from any [Worker](https://flaviocopes.com/cloudflare-workers/):

```jsonc
{
  "ai": {
    "binding": "AI"
  }
}
```

Small specialized models running at the edge, milliseconds from users, is exactly the direction I expected Workers AI to go. A vision model that answers "where is the button in this screenshot" in under 150 ms opens more doors than another chat model would.