Moondream3 Preview Caption

Moondream3 Preview Caption

Playground

Generate short, normal, or long image captions to help you understand and describe visual content. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Moondream 3 Caption is a high-performance vision-language model that automatically generates clear, descriptive, and context-aware captions for any image. It supports multiple caption lengths, enabling flexible use across social media content, dataset annotation, and creative storytelling.

✨ Key Features

• Flexible Caption Length Choose from short, normal, or long captions to fit your workflow needs.

• Accurate Visual Understanding Trained on large-scale, diverse visual datasets — accurately detects objects, actions, and environments.

• Fast and Efficient Optimized for low-latency inference, suitable for real-time applications and batch processing.

• Human-like Language Output Produces smooth, natural, and grammatically correct sentences ideal for direct use in production.

⚙️ Example Usage

🔹 Short Caption

{
 "image": "https://example.com/photo.jpg",
 "length": "short"
}

🔹 Normal Caption

{
 "image": "https://example.com/photo.jpg",
 "length": "normal"
}

🔹 Long Caption

{
 "image": "https://example.com/photo.jpg",
 "length": "long"
}

🧾 Example Output

{
 "caption": "A young woman with long, dark hair stands in front of a bar. She wears a leopard print halter top and blue jeans, accessorized with large hoop earrings. The bar features a purple backlit counter and a lit sign displaying 'DAMON' in yellow letters."
}

Output Explanation

• The model returns a JSON object with a single key: caption.
• The value is a natural-language description automatically generated from the input image.
• The style and length of the caption depend on your chosen length parameter (short, normal, or long).

💡 Best Practices

• Use “short” for quick summaries or thumbnail text.
• Use “normal” for descriptive captions (recommended default).
• Use “long” for storytelling, research annotations, or dataset labeling.
• Supported formats: JPEG, PNG, WebP
• Maximum image size: 10 MB

💰 Pricing

• $0.005 per request
• Contact WaveSpeedAI for enterprise and large-scale pricing options.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "length": "normal"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/moondream3-preview/caption" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters