> For the complete documentation index, see [llms.txt](https://docs.roboflow.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.roboflow.com/roboflow/roboflow-hi/deploy/supported-models/moondream2.md).

# Moondream2

Moondream2 एक कॉम्पैक्ट vision-language model है। Roboflow Inference में, इसे एक open-vocabulary object detector के रूप में exposed किया जाता है: एक class name को prompt के रूप में pass करें और matching regions के लिए bounding boxes प्राप्त करें।

{% hint style="info" %}
Moondream2 Serverless Hosted API पर उपलब्ध नहीं है। इसे एक पर चलाएँ [Dedicated Deployment](/roboflow/roboflow-hi/deploy/dedicated-deployments.md) या [self-hosted Inference](https://inference.roboflow.com/).
{% endhint %}

## कोड नमूना

{% stepper %}
{% step %}

### अपनी API Key प्राप्त करें

एक Roboflow खाता बनाएं, अपनी key यहाँ पर ढूँढें [Roboflow API settings page](https://app.roboflow.com/settings/api) और इसे अपने shell में उपलब्ध कराएँ:

```bash
export ROBOFLOW_API_KEY="your-key-here"
```

{% endstep %}

{% step %}

### निर्भरताएँ इंस्टॉल करें

इंस्टॉल करें [Inference SDK](https://inference.roboflow.com/) और [supervision](https://supervision.roboflow.com/):

```bash
pip install inference-sdk supervision opencv-python
```

{% endstep %}

{% step %}

### मॉडल चलाएँ

सेट करें `api_url` को अपने Dedicated Deployment URL या local Inference server पर।

```python
import os
import cv2
import numpy as np
import requests
import supervision as sv
from inference_sdk import InferenceHTTPClient

content = requests.get("https://media.roboflow.com/notebooks/examples/dog.jpeg").content
image = cv2.imdecode(np.frombuffer(content, np.uint8), cv2.IMREAD_COLOR)
client = InferenceHTTPClient(
    api_url="https://your-deployment.roboflow.cloud",
    api_key=os.environ["ROBOFLOW_API_KEY"],
)
result = client.infer_lmm(
    image,
    model_id="moondream2",
    prompt="dog",
)

preds = result["predictions"]
xyxys = [
    [p["x"] - p["width"] / 2, p["y"] - p["height"] / 2,
     p["x"] + p["width"] / 2, p["y"] + p["height"] / 2]
    for p in preds
]
detections = sv.Detections(
    xyxy=np.array(xyxys, dtype=float),
    class_id=np.array([p.get("class_id", 0) for p in preds]),
    confidence=np.array([p.get("confidence", 1.0) for p in preds], dtype=float),
    data={"class_name": np.array([p["class"] for p in preds])},
)
labels = [f"{p['class']} {p.get('confidence', 1.0):.2f}" for p in preds]
annotated = sv.BoxAnnotator().annotate(image.copy(), detections)
annotated = sv.LabelAnnotator().annotate(annotated, detections, labels=labels)
cv2.imwrite("dog_annotated.png", annotated)
```

<figure><img src="/files/7abbd76a10caf638c8bedfe2d66b5735700c9ab7" alt=""><figcaption></figcaption></figure>
{% endstep %}
{% endstepper %}

## Inference speed

Latency मापी गई [Roboflow Inference](https://inference.roboflow.com/) 1x NVIDIA L4 पर, batch size 1 के साथ, एक image की captioning करते समय। Moondream2 अपने output length को fix नहीं कर सकता, इसलिए latency response के साथ बदलती रहती है।

<table data-search="false"><thead><tr><th>उपनाम</th><th>विलंबता (ms)</th></tr></thead><tbody><tr><td><code>moondream2</code></td><td>1669</td></tr></tbody></table>

{% hint style="info" %}
सेट करें `api_url` को अपने deployment target से मिलाएँ:

* `http://localhost:9001` एक local [Inference](https://inference.roboflow.com/) server.
* आपका [Dedicated Deployment](/roboflow/roboflow-hi/deploy/dedicated-deployments.md) एक private endpoint के लिए URL.
  {% endhint %}