Google Imagen3 API Documentation

Google Imagen3 API Documentation

Playground

Imagen3 is Google’s highest-quality text-to-image model, generating highly detailed, beautifully lit and photoreal images from text prompts. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Generate premium images with Google Imagen 3 — Google’s most advanced text-to-image model. Renowned for exceptional photorealism, nuanced understanding of prompts, and superior detail rendering, Imagen 3 delivers the highest quality output for professional creative work.

Looking for faster generation? Try Google Imagen 3 Fast for speed-optimized output at a lower price.

Why It Looks Great

• Google’s best: The flagship Imagen model with maximum quality and detail.
• Superior photorealism: Exceptional at lifelike scenes, portraits, and natural imagery.
• Nuanced understanding: Accurately interprets complex, detailed prompts.
• Batch support: Generate up to 4 images per request.
• Negative prompts: Exclude unwanted elements for precise control.
• Flexible aspect ratios: Multiple format options for any use case.
• Prompt Enhancer: Built-in tool to refine your descriptions automatically.
• Reproducible results: Use the seed parameter to recreate exact outputs.

Parameters

How to Use

1. Write your prompt — describe the image you want to create in detail.
2. Use Prompt Enhancer (optional) — click to automatically enrich your description.
3. Set aspect ratio — choose the format that fits your needs.
4. Choose number of images — generate 1-4 variations at once.
5. Add negative prompt (optional) — specify elements to exclude.
6. Set seed (optional) — for reproducible results.
7. Run — click the button to generate.
8. Download — preview and save your images.

Pricing

Flat rate per image generated.

Best Use Cases

• Professional Photography — Generate photorealistic images for commercial use.
• Marketing & Advertising — Create high-quality visuals for campaigns.
• Product Visualization — Produce premium product and lifestyle imagery.
• Editorial Content — Generate publication-quality illustrations and photos.
• Final Deliverables — Premium output suitable for client-facing work.

Example Prompts

• “Professional portrait of a chef in a modern kitchen, natural lighting, genuine smile, editorial quality”
• “Luxury watch product shot on marble surface, dramatic lighting, reflection details”
• “Family picnic in a sun-dappled park, candid moment, warm afternoon light, lifestyle photography”
• “Architectural interior of a minimalist Scandinavian living room, soft natural light, magazine quality”
• “Close-up of fresh sushi platter, restaurant lighting, appetizing presentation, food photography”

Model Comparison

Pro Tips for Best Results

• Use Imagen 3 for final deliverables where quality is paramount.
• Include photography terms: “editorial quality”, “magazine quality”, “professional lighting”.
• Batch generate (num_images: 4) to get variations and pick the best.
• Test concepts with Imagen 3 Fast, then generate finals with standard Imagen 3.
• Negative prompts help avoid: “blur”, “artifacts”, “distortion”, “watermark”.
• Detailed prompts yield better results — describe lighting, mood, and style.

Notes

• Maximum 4 images per generation request.
• Imagen 3 delivers Google’s highest quality image generation.
• Processing takes longer than Fast mode but produces superior results.
• Ideal for professional and commercial creative work.

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": "1:1",
  "num_images": 1
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/google/imagen3" \
  -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