Krea V2 Turbo API Documentation

Krea V2 Turbo API Documentation

Playground

Krea 2 Turbo generates aesthetic-first images natively at 1K to 2K resolution in seconds, supporting both text-to-image and image-to-image. Ready-to-use REST inference API, best performance, no cold starts, affordable pricing.

Features

Krea 2 Turbo is a fast image generation model built for aesthetic-first text-to-image and image-to-image workflows. The Turbo checkpoint is distilled for fast generation while supporting native output from 1k up to 2k resolution, making it suitable for both quick iteration and higher-resolution final assets.

Why Choose This?

• Native 2K generation
  Generate higher-resolution images directly without requiring a separate upscaling step.

• Fast Turbo workflow
  Use the distilled Turbo checkpoint for faster image generation and prompt iteration.

• Aesthetic-first output
  Designed for photographic realism, strong art direction, and polished visual style.

• Text-to-image and image-to-image
  Generate from text prompts or provide an image URL for image-to-image variation and restyling.

• Strength control
  Use strength to control how strongly the model repaints the source image in image-to-image mode.

• Flexible aspect ratios
  Supports square, portrait, landscape, tall, wide, and cinematic aspect ratio presets.

Parameters

How to Use

1. Write your prompt — Describe the subject, style, lighting, composition, and visual details.
2. Choose resolution — Use 1k for faster iteration or 2k for higher-resolution output.
3. Select aspect ratio — Pick the layout that matches your target format.
4. Add image optional — Provide an image URL when using image-to-image mode.
5. Adjust strength optional — Lower values preserve more of the source image; higher values repaint more freely.
6. Set seed optional — Use a fixed seed for more reproducible results.
7. Submit — Generate the final image output.

Pricing

Pricing is based on selected resolution.

Best Use Cases

• Fast image generation — Create polished images from text prompts.
• Prompt iteration — Test multiple creative directions at lower cost using 1k.
• Final image assets — Use 2k when higher-resolution output is needed.
• Image variation — Restyle or reinterpret an existing image with prompt control.
• Marketing visuals — Generate campaign images, social posts, product concepts, and creative assets.
• Concept art — Explore characters, environments, scenes, and visual moods.

Pro Tips

• Use detailed prompts with subject, composition, lighting, camera, mood, and style direction.
• Use 1k for quick iteration and 2k for final assets.
• For image-to-image, use lower strength values to preserve composition and higher values for stronger restyling.
• Set a fixed seed when comparing prompt or parameter changes.
• Choose aspect ratios based on the target platform, such as 9:16 for vertical content or 16:9 for widescreen layouts.

Related Models

• Krea 2 Turbo LoRA — Generate images with the same Turbo workflow and LoRA support.

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",
  "resolution": "1k",
  "strength": 0.8
}
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/wavespeed-ai/krea-v2/turbo" \
  -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