Browse Models

API Integration

WaveSpeed AI submits tasks through /api/v3/{model_id}. Use the complete model ID shown on the model page, for example wavespeed-ai/z-image/turbo; depending on the model, the ID can contain either two segments (vendor/model) or three (vendor/model/variant). Add Authorization: Bearer <YOUR_API_KEY> and Content-Type: application/json headers to every request.

Example Usage


curl --fail-with-body --connect-timeout 10 --max-time 60 \
--request POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/z-image/turbo" \
--header "Authorization: Bearer $WAVESPEED_API_KEY" \
--header "Content-Type: application/json" \
--data-raw '{
  "prompt": "Octopus vs crab chess game, underwater, vibrant colors",
  "size": "1024*1024"
}'

Do not blindly retry task-submission POST requests: a disconnected response can still correspond to a prediction that was accepted and billed. Retry idempotent result-query GET requests instead, using the returned prediction ID or data.urls.get.

© 2026 WaveSpeedAI. All rights reserved.