Z Image Turbo Image To Image LoRA API Documentation

Z Image Turbo Image To Image LoRA API Documentation

Playground

Z-Image-Turbo Image-to-Image LoRA transforms reference images with custom LoRA styles in sub-second time. Apply up to 3 LoRAs for personalized image transformation. Ready-to-use REST inference API, best performance, no cold starts, affordable pricing.

Features

Z-Image Turbo Image-to-Image LoRA is a versatile image generation model with full LoRA support. Apply up to 3 custom LoRA adapters while controlling the transformation spectrum — from subtle enhancement to dramatic reimagination — all via the strength parameter.

This is not a simple editing tool — it’s a generation engine that combines custom styles with flexible image modification.

Looking for the standard version? Try Z-Image Turbo Image-to-Image without LoRA support.

Why Choose This?

• LoRA support Apply up to 3 custom LoRA adapters to personalize style, characters, or visual aesthetics.

• Flexible transformation spectrum From near-lossless enhancement to complete style overhaul — all controlled by a single strength slider.

• Quality enhancement mode At low strength, improve image quality and add detail while applying subtle LoRA styles.

• Creative reimagination mode At high strength, combine LoRA styles with dramatic image transformation.

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

Understanding Strength

The strength parameter controls how much the output differs from your input image:

Parameters

LoRA Format

Each LoRA in the loras array has:

• path (required) — URL to the LoRA weights file
• scale (optional) — Weight multiplier, default 1

How to Use

1. Upload your image — drag and drop or paste a public URL.
2. Write your prompt — describe the desired output or transformation.
3. Add LoRAs — click ”+ Add Item” to include custom LoRA adapters (up to 3).
4. Set strength — low for enhancement, high for reimagination.
5. Adjust output size — set width and height as needed.
6. Run — submit and download the transformed image.

Pricing

Simple flat-rate pricing regardless of image size, strength, or LoRA count.

Best Use Cases

• Character Transformation — Apply character LoRAs while preserving pose and composition.
• Style Transfer with Control — Use style LoRAs with adjustable strength for precise blending.
• Brand Consistency — Transform images to match brand aesthetics via custom LoRAs.
• Artistic Reinterpretation — Combine multiple art style LoRAs for unique hybrid looks.
• Quality Enhancement — Improve image quality while subtly applying trained styles.

Pro Tips

• Start with strength 0.5-0.6 to balance original image and LoRA influence.
• LoRA effect is more visible at higher strength values.
• Combine LoRAs carefully — multiple LoRAs can conflict; test combinations.
• If your LoRA uses trigger words, include them in your prompt.
• Use the same seed to compare different strength levels or LoRA combinations.

Notes

• Up to 3 LoRAs can be applied simultaneously.
• LoRA version pricing is 2× the standard version.
• Output dimensions can differ from input — set width and height to your target size.
• At low strength, the original image dominates; at high strength, LoRA styles dominate.

Related Models

• Z-Image Turbo Image-to-Image — Standard version without LoRA support.
• Z-Image Turbo Text-to-Image — Generate images from text prompts only.
• Z-Image LoRA Trainer — Train custom LoRAs for Z-Image models.

Reference

• How to train your LoRA?
• How to use your LoRA?

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",
  "size": "1024*1024",
  "strength": 0.6,
  "seed": -1,
  "output_format": "jpeg"
}
JSON
)

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