Runwayml Gen4 Image Turbo API Documentation

Runwayml Gen4 Image Turbo API Documentation

Playground

RunwayML Gen-4 Image Turbo is cheaper and 2.5x faster than Gen-4 Image and supports up to 3 reference images to capture every angle. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Generate stunning images from text prompts with optional reference image guidance. Runway’s Gen4 Image Turbo delivers fast, high-quality results with support for multiple aspect ratios and resolutions — perfect for rapid creative iteration.

Why It Looks Great

• Reference-guided generation: Upload up to 3 reference images to guide style, composition, or subject appearance.
• Flexible aspect ratios: Supports multiple formats including 9:16, 16:9, 1:1, and more for any use case.
• High resolution output: Generate images up to 1080p for crisp, detailed results.
• Fast turbo processing: Optimized for speed without sacrificing quality.
• Reproducible results: Use the seed parameter to recreate exact outputs or explore variations.

Parameters

How to Use

1. Write your prompt — describe the image you want to create in detail.
2. Choose aspect ratio — select the format that fits your needs (9:16 for portraits/stories, 16:9 for landscapes, etc.).
3. Select resolution — pick your desired output quality.
4. Add reference images (optional) — upload up to 3 images to guide style, subject, or composition.
5. Set seed (optional) — use a specific number for reproducible results, or leave empty for random.
6. Run — click the button to generate.
7. Download — preview and save your generated image.

Pricing

Flat rate of $0.035 per generated image.

Best Use Cases

• Social Media Content — Create scroll-stopping visuals for Instagram, TikTok, and stories with 9:16 aspect ratio.
• Concept Art & Ideation — Rapidly explore visual ideas and creative directions.
• Marketing & Advertising — Generate on-brand imagery for campaigns and promotions.
• Character Design — Use reference images to maintain consistent character appearance across generations.
• Style Transfer — Apply the aesthetic of reference images to new scenes and subjects.

Pro Tips for Best Results

• Be specific in your prompts — include details about lighting, mood, style, and composition.
• Use reference images strategically: one for style, one for subject, one for composition.
• Keep the seed fixed when iterating on prompts to isolate the effect of text changes.
• For consistent characters or styles across multiple images, always include the same reference image.
• Match aspect ratio to your intended use case: 9:16 for mobile/stories, 16:9 for desktop/video thumbnails, 1:1 for profile pictures.

Notes

• If using URLs for reference images, ensure they are publicly accessible. A preview thumbnail confirms successful loading.
• The enable_base64_output option is only available through the API, not the web interface.
• Reference images influence but do not guarantee exact replication — they guide style and composition.

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'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "aspect_ratio": "4:3",
  "resolution": "1080p"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/runwayml/gen4-image-turbo" \
  -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