Kwaivgi Kling Image V3 Edit API Documentation

Kwaivgi Kling Image V3 Edit API Documentation

Playground

Kling V3 Edit is an AI model for editing and transforming images via text prompts, enabling precise modifications with natural-language instructions. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Kling Image V3 Edit is Kuaishou’s image editing model that transforms existing images based on text instructions. Upload a reference image and describe the changes you want — the model applies edits while preserving the original style, structure, and identity. Supports flexible aspect ratios, resolution options, and batch generation.

Why Choose This?

• Text-guided editing Describe changes in natural language — no manual masking or layer editing required.

• Style preservation Maintains the original image’s composition, lighting, and aesthetic while applying your edits.

• Flexible aspect ratios Multiple options including 1:1, 3:4, 4:3, 9:16, 16:9 and more to fit any use case.

• Resolution control Choose output resolution (1k and above) based on your quality and speed requirements.

• Batch generation Generate multiple variations in a single request for rapid iteration and comparison.

• Prompt Enhancer Built-in tool to automatically improve your edit descriptions for better results.

Parameters

How to Use

1. Upload your image — provide the reference image you want to edit.
2. Write your prompt — describe the changes you want (e.g., “Change the jacket to red leather”).
3. Choose aspect ratio — select the output format that fits your use case.
4. Set resolution — choose 1k for speed or higher for more detail.
5. Set num_images — generate multiple variations if needed.
6. Run — submit and download your edited images.

Pricing

Billing Rules

• Rate: $0.028 per image
• Total cost = num_images × $0.028

Best Use Cases

• Fashion & E-commerce — Change clothing colors, styles, or patterns on product images.
• Creative Iteration — Explore variations of a concept without starting from scratch.
• Marketing & Ads — Adapt visuals for different campaigns while maintaining brand consistency.
• Portrait Editing — Adjust styling, backgrounds, or accessories in portrait photos.
• Content Repurposing — Transform existing images for new contexts or platforms.

Pro Tips

• Be specific about what should change and what should stay the same.
• Use the Prompt Enhancer to refine vague instructions into detailed edit commands.
• Generate multiple images (num_images > 1) to explore different interpretations of your edit.
• Higher quality source images yield better editing results.
• Use png format when you need lossless quality output.

Notes

• Both prompt and image are required fields.
• Ensure uploaded image URLs are publicly accessible.
• Higher resolution may slightly increase processing time.

Related Models

• Kling Image V3 Text-to-Image — Generate images from text prompts.
• Kling Image O3 Text-to-Image — Next-generation O3 architecture for text-to-image.
• Kling Video O3 Pro Image-to-Video — Animate edited images into video.

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",
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "aspect_ratio": "16:9",
  "resolution": "1k",
  "num_images": 1,
  "output_format": "png"
}
JSON
)

# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/kwaivgi/kling-image-v3/edit" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}"); then
  printf 'Submit failed. %s\n' "${CHECK}" >&2
  exit 1
fi

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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
  # Transient errors (429, 5xx) are retried, honoring Retry-After.
  HTTP_CODE=$(curl --silent --show-error --fail-with-body \
    --connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
    --output "${BODY_FILE}" --write-out '%{http_code}' \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
  case "${HTTP_CODE}" in
    2??) ;;
    000|429|5??)  # dropped connection or temporary error: keep polling
      printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      sleep 10; continue ;;
    *) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      cat "${BODY_FILE}" >&2; exit 1 ;;
  esac
  RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted)
      printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
      printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

Result Request Parameters

Result Response Parameters