Krea V2 Large Text To Image API Documentation

Krea V2 Large Text To Image API Documentation

Playground

Krea 2 Large Text to Image is a fast AI image generation model that creates high-fidelity images from text prompts with aspect ratio, creativity, and optional style reference controls. Ready-to-use REST inference API for creative design, marketing visuals, product mockups, brand assets, social media content, concept art, and professional text-to-image workflows with simple integration, no coldstarts, and affordable pricing.

Features

WaveSpeed AI Krea V2 Large Text-to-Image generates high-quality images from natural-language prompts, with optional reference images for stronger style guidance. It is designed for premium prompt-based image generation workflows where you want flexible aspect ratios, controllable creativity, and the option to steer the final look with one or more visual references.

Why Choose This?

• High-quality text-to-image generation
  Create polished images from detailed natural-language prompts.

• Reference-guided style control
  Add up to 10 reference images to guide the visual style of the generated result.

• Flexible aspect ratios
  Choose from multiple preset aspect ratios for square, portrait, landscape, or cinematic compositions.

• Creativity control
  Adjust how loosely the model interprets the prompt with raw, low, medium, or high.

• Multi-reference support
  Use multiple references with individual strength values for more nuanced style influence.

• Production-ready workflow
  Suitable for concept art, marketing visuals, brand creatives, editorial imagery, and visual ideation.

Parameters

Reference Format

Each item in the reference array supports:

How to Use

1. Write your prompt — describe the subject, style, lighting, composition, and mood you want.
2. Choose aspect ratio (optional) — select the format that best fits your target use case.
3. Set creativity (optional) — use raw or low for tighter prompt control, or medium / high for looser interpretation.
4. Add reference images (optional) — upload one or more images if you want stronger style guidance.
5. Adjust reference strength (optional) — tune each reference image’s influence individually.
6. Submit — run the model and download the generated image.

Example Prompt

A premium editorial portrait of a woman in soft window light, natural skin texture, elegant neutral wardrobe, cinematic depth of field, refined luxury-magazine styling

Pricing

Pricing is based on whether you use reference images.

Billing Rules

• Base price is $0.06 per image
• Adding reference images adds $0.005 to the request
• The number of reference images does not change the surcharge beyond that single addition
• aspect_ratio and creativity do not affect pricing
• Each request returns one generated image

Best Use Cases

• Prompt-based concept generation — Explore visual directions from text prompts alone.
• Style-guided image creation — Use reference images to steer the output toward a desired aesthetic.
• Editorial and fashion visuals — Build polished, art-directed imagery with better style control.
• Brand creative development — Keep generated visuals closer to an existing visual language.
• Marketing and campaign ideation — Generate premium-looking assets for pitches, mockups, and content planning.

Pro Tips

• Use raw or low creativity when you want stricter prompt fidelity.
• Use medium or high when you want more interpretive or stylized output.
• Add reference images only when you want style guidance, since they increase the price slightly.
• Start with a single strong reference before stacking multiple references.
• Adjust strength carefully when combining several references, especially if their styles differ.
• Be specific in your prompt about subject, lighting, composition, and mood for more controllable results.

Notes

• prompt is required.
• reference is optional and supports up to 10 images.
• Each reference image can use a strength value from -2 to 2.
• creativity defaults to medium.
• aspect_ratio defaults to 1:1.
• Pricing is fixed per request, with a small surcharge when reference images are used.

Related Models

• WaveSpeed AI Krea V2 Large Image-to-Image — Edit or restyle an existing image with Krea V2 Large.
• WaveSpeed AI FLUX.2 Klein Base 9B Text-to-Image — Generate images from prompts with a simpler parameter set.
• WaveSpeed AI FLUX.2 Klein Base 9B Text-to-Image LoRA — Generate images from prompts with custom 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",
  "creativity": "medium"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/krea-v2-large/text-to-image" \
  -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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 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|deleted) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters