# Moondream3 Preview [Caption] API

> Viết chú thích cho hình ảnh với độ dài ngắn, bình thường hoặc dài

Nhận hình ảnh từ `image_url` và trả về chú thích dạng văn bản. Tùy chọn `length` chọn độ dài short, normal hoặc long, trong đó normal là mặc định. `temperature` và `top_p` tùy chọn, đều từ 0 đến 1, điều khiển việc lấy mẫu. Đầu ra gồm chú thích, lý do kết thúc và thông tin sử dụng.

- Model id: `moondream3-preview/caption`
- Made by: Moondream
- Category: Vision
- Output: text
- Estimated price per run: $0.002 / run (cheapest run $0.002 (2 API credits))
- Run: `POST https://api.doitong.com/v1/run/moondream3-preview/caption`
- Quote: `POST https://api.doitong.com/v1/quote/moondream3-preview/caption`
- Docs page: https://doitong.com/vi/developers/models/moondream3-preview/caption

## Pricing

Price: **$0.002 / run**. Estimated from the typical cost of one run. The quote is exactly what the run charges.

The full example input below costs $0.002 = 2 API credits (1 request × 2 API credits).

Prices are the upstream price in USD, paid from the API balance in API credits (1000 API credits = 1 USD). `POST https://api.doitong.com/v1/quote/moondream3-preview/caption` with the same JSON body returns the exact price for that input; the quote is exactly what a run charges. Invalid input is never charged and failed or canceled requests are refunded automatically.

## Input parameters

Send either the canonical name or the native field name (not both). Canonical names are the same across models.

| Parameter | Canonical name | Type | Required | Default | Allowed values / range | Description |
|---|---|---|---|---|---|---|
| `image_url` | `image` | string (image URL) | yes |  |  | URL trỏ đến hình ảnh cần xử lý. |
| `length` |  | enum (string) | no | `normal` | `short`, `normal`, `long` | Độ dài mong muốn của chú thích được tạo. |
| `temperature` |  | number \| null | no |  | 0 – 1 | Nhiệt độ lấy mẫu, nằm trong khoảng từ 0 đến 1; mặc định là 0 khi bị bỏ qua. Giá trị cao hơn như 0.8 làm tăng tính ngẫu nhiên, còn giá trị thấp hơn như 0.2 tạo đầu ra tập trung và xác định hơn. |
| `top_p` |  | number \| null | no |  | 0 – 1 | Khối lượng xác suất cho lấy mẫu nucleus, nằm trong khoảng từ 0 đến 1. |

File parameters (`image`) take a public `https://` URL, a `data:` URI up to 10 MB, or the `file_url` of an upload from `POST /files`.

## Example input

Required parameters only:

```json
{
  "image": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
}
```

A fuller example:

```json
{
  "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
}
```

## Run it

Set `DOITONG_API_KEY` to a key from https://doitong.com/developers/dashboard#keys. `?wait=20` returns the finished request in the same call when it is ready within 20 seconds; otherwise poll `GET /requests/{id}` (it also accepts `?wait=`), or pass `webhook_url`.

### curl

```bash
# Run the model; wait up to 20 s for the finished result in the same call
curl -X POST "https://api.doitong.com/v1/run/moondream3-preview/caption?wait=20" \
  -H "Authorization: Bearer $DOITONG_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
  }'

# Still "queued" or "running"? Poll with the id from the response:
curl "https://api.doitong.com/v1/requests/REQUEST_ID?wait=20" \
  -H "Authorization: Bearer $DOITONG_API_KEY"
```

### JavaScript

```javascript
const API = "https://api.doitong.com/v1";
const headers = {
  "Authorization": "Bearer " + process.env.DOITONG_API_KEY,
  "Content-Type": "application/json",
};

// Run the model; wait up to 20 s for the finished result in the same call
let res = await fetch(API + "/run/moondream3-preview/caption?wait=20", {
  method: "POST",
  headers,
  body: JSON.stringify({
    "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
  }),
});
let request = await res.json();
if (!res.ok) throw new Error(request.error.code + ": " + request.error.message);

// Not finished yet: long-poll until it is
while (request.status === "queued" || request.status === "running") {
  res = await fetch(API + "/requests/" + request.id + "?wait=20", { headers });
  request = await res.json();
}
if (request.status !== "succeeded") throw new Error(JSON.stringify(request.error));
console.log(request.outputs.map((o) => o.url), request.cost.credits + " API credits ($" + request.cost.usd + ")");
```

### Python

```python
import os
import requests

API = "https://api.doitong.com/v1"
headers = {"Authorization": "Bearer " + os.environ["DOITONG_API_KEY"]}

payload = {
    "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg",
}

# Run the model; wait up to 20 s for the finished result in the same call
r = requests.post(API + "/run/moondream3-preview/caption", params={"wait": 20}, json=payload, headers=headers)
request = r.json()
if not r.ok:
    raise RuntimeError(request["error"]["code"] + ": " + request["error"]["message"])

# Not finished yet: long-poll until it is
while request["status"] in ("queued", "running"):
    request = requests.get(API + "/requests/" + request["id"], params={"wait": 20}, headers=headers).json()
if request["status"] != "succeeded":
    raise RuntimeError(request["error"])
print([o["url"] for o in request["outputs"]], request["cost"]["credits"], "API credits", "$" + request["cost"]["usd"])
```

### MCP

The MCP server at `https://api.doitong.com/v1/mcp` exposes the tools `search_models`, `get_model`, `quote`, `run`, `get_request`, `cancel_request`, `list_requests`, `create_upload` and `get_account`. A `run` call for this model:

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "run",
    "arguments": {
      "model_id": "moondream3-preview/caption",
      "input": {
        "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
      },
      "wait_seconds": 20
    }
  }
}
```

## Response

Every call returns a request object. `status` is one of `queued`, `running`, `succeeded`, `failed`, `canceled`. `outputs` lists the generated files with URLs on our storage (kept forever, or 7 days with `?storage=temporary`; `storage` says which); `output` is the model's own result object with the same URLs.

```json
{
  "id": "req_8k2m4q7x1z9c3v5b6n0p2r4t",
  "object": "request",
  "model": "moondream3-preview/caption",
  "status": "succeeded",
  "created_at": "2026-10-20T12:00:00Z",
  "started_at": "2026-10-20T12:00:02Z",
  "completed_at": "2026-10-20T12:00:09Z",
  "queue_position": null,
  "cost": {
    "credits": 2,
    "usd": "0.002",
    "charged": 2,
    "refunded": 0
  },
  "input": {
    "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
  },
  "output": {
    "finish_reason": "stop",
    "usage_info": {
      "input_tokens": 1,
      "output_tokens": 1,
      "prefill_time_ms": 1.0,
      "decode_time_ms": 1.0,
      "ttft_ms": 1.0
    },
    "output": "A hedgehog is captured in a close-up shot, focusing on its face and nose. The hedgehog's spines are visible along its back, and its nose is dark and wet-looking. A gold ring with a small diamond is positioned on the grass in front of the hedgehog, partially obscured by its nose. The background is a blurred green grassy field, with small white flowers scattered throughout the grass."
  },
  "outputs": [
    {
      "type": "file",
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.jpeg",
      "content_type": "image/jpeg",
      "file_size": 1843200,
      "stored": true
    }
  ],
  "error": null,
  "metrics": {
    "inference_seconds": 6.8
  },
  "urls": {
    "self": "https://api.doitong.com/v1/requests/req_8k2m4q7x1z9c3v5b6n0p2r4t",
    "cancel": "https://api.doitong.com/v1/requests/req_8k2m4q7x1z9c3v5b6n0p2r4t/cancel"
  }
}
```

### Output fields (`output`)

| Field | Type | Description |
|---|---|---|
| `finish_reason` | string | Lý do quá trình tạo đầu ra kết thúc. |
| `usage_info` | object {input_tokens, output_tokens, prefill_time_ms, decode_time_ms, ttft_ms} | Chi tiết mức sử dụng của yêu cầu. |
| `output` | string | Chú thích được tạo để mô tả hình ảnh. |

## Related models

- [Moondream3 Preview \[Segment\]](https://doitong.com/developers/models/moondream3-preview/segment.md) (`moondream3-preview/segment`): $0.003 / run
- [Moondream 3 Preview \[Query\]](https://doitong.com/developers/models/moondream3-preview/query.md) (`moondream3-preview/query`): $0.001 / run
- [Moondream3 Preview \[Detect\]](https://doitong.com/developers/models/moondream3-preview/detect.md) (`moondream3-preview/detect`): $0.001 / run
- [Moondream3 Preview \[Point\]](https://doitong.com/developers/models/moondream3-preview/point.md) (`moondream3-preview/point`): $0.001 / run
- [Moondream](https://doitong.com/developers/models/moondream/batched.md) (`moondream/batched`): $0.004 / run
- [Moondream2](https://doitong.com/developers/models/moondream2.md) (`moondream2`): $0.01 / run
- [Moondream2 · Object Detection](https://doitong.com/developers/models/moondream2/object-detection.md) (`moondream2/object-detection`): $0.02 / image
- [Moondream2 · Point Object Detection](https://doitong.com/developers/models/moondream2/point-object-detection.md) (`moondream2/point-object-detection`): $0.02 / image

## Links

- HTML page: https://doitong.com/developers/models/moondream3-preview/caption
- All models and guides for AI agents: https://doitong.com/developers/llms.txt
- OpenAPI 3.1 with this model's typed request body: https://doitong.com/developers/openapi.json?model=moondream3-preview/caption
- Errors: https://doitong.com/developers/docs/errors
- Model details as JSON: https://api.doitong.com/v1/models/moondream3-preview/caption