Molmo2 Prompt Optimizer API Documentation

Molmo2 Prompt Optimizer API Documentation

Playground

Molmo2-4B Prompt Optimizer: Enhance prompts for image and video generation with intelligent restructuring, style guidance, and context-aware improvements. Open-source vision-language model. Ready-to-use REST API, no cold starts, affordable pricing.

Features

Enhance your prompts for better AI generation results with Molmo2 Prompt Optimizer. This intelligent tool analyzes your input — whether text or image — and generates optimized prompts tailored for different styles and output modes. Perfect for improving text-to-image and text-to-video generation quality.

Why It Works Great

• Dual input modes: Analyze images or text to generate optimized prompts.
• Multiple styles: Optimize for artistic, photographic, technical, anime, or realistic output.
• Image & video support: Generate prompts optimized for either image or video generation.
• Instant results: Fast processing for seamless workflow integration.
• Ultra-affordable: Just $0.003 per optimization — 333 runs for $1.
• Quality boost: Get better results from your generation models.

Parameters

How to Use

From Image

1. Upload an image — the image you want to describe or recreate.
2. Select style — choose the aesthetic direction.
3. Select mode — image or video generation target.
4. Run — get an optimized prompt describing the image.

From Text

1. Enter your prompt — your basic idea or description.
2. Select style — choose the aesthetic direction.
3. Select mode — image or video generation target.
4. Run — get an enhanced, detailed prompt.

Pricing

Flat rate per optimization.

Style Guide

Mode Options

Best Use Cases

• Image Captioning — Generate detailed prompts from reference images.
• Prompt Enhancement — Upgrade basic prompts to detailed descriptions.
• Style Conversion — Reframe prompts for different aesthetic styles.
• Cross-model Optimization — Adapt prompts for image vs video generation.
• Batch Processing — Optimize many prompts affordably at scale.
• Learning Tool — Understand what makes effective generation prompts.

Pro Tips for Best Results

• Use image input to reverse-engineer prompts from existing artwork.
• Match style to your target model’s strengths (anime for anime models, etc.).
• Switch mode to “video” when targeting video generation for motion-aware prompts.
• Combine: upload image + add text for context-aware optimization.
• At $0.003 per run, experiment freely to find optimal prompt styles.

Notes

• Provide either image or text (or both) as input.
• Processing is near-instant for rapid iteration.
• Video mode adds motion and temporal descriptions to prompts.
• Results can be directly used with generation models on WaveSpeed.

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'
{
  "style": "default",
  "mode": "image"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/molmo2/prompt-optimizer" \
  -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