Kwaivgi Kling V3 Turbo Pro Text To Video API Documentation

Kwaivgi Kling V3 Turbo Pro Text To Video API Documentation

Playground

Kling V3 Turbo Pro converts text prompts into high-quality 1080P videos with fast pro-tier inference, single-prompt generation, and multi-shot storyboard support. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Kling V3 Turbo Pro Text-to-Video generates high quality 1080p videos from text prompts. It supports both single-prompt generation and multi-shot storyboard generation with per-shot durations.

Why Choose This?

• High quality 1080p video generation
  Create polished pro-tier videos directly from text prompts.

• Fast turbo inference
  Use the turbo version for faster generation while keeping strong visual quality.

• Single prompt or multi-shot storyboard
  Generate a complete video from one prompt, or use multi_prompt to define multiple shots.

• Per-shot duration control
  Set each storyboard segment duration when using multi_prompt.

• Standard video output
  The generated video is returned as a URL in the standard WaveSpeed prediction response.

Parameters

How to Use

1. Enter a prompt — Describe the scene, subject, motion, and style you want.
2. Or create a storyboard — Use multi_prompt to define multiple shots with individual durations.
3. Choose aspect ratio — Select 16:9, 9:16, or 1:1.
4. Set duration — Choose the total video duration. For multi_prompt, set each shot duration so their sum equals the selected total.
5. Submit — Generate the video output.

Output

Returns generated video URL(s) in the standard WaveSpeed prediction response.

The generated video is returned as MP4 video.

Pricing

Pricing is $0.14 per second.

WaveSpeed bills by the selected total duration. For multi_prompt, shot durations must sum to that total; they are not added again for billing.

Billing Rules

• Billing is based on generated video duration.
• The duration parameter is the total generated video duration.
• For multi_prompt, shot durations must sum to duration; billing does not add them a second time.
• Each second costs $0.14.

Best Use Cases

• Text-to-video creation — Turn written concepts into finished video clips.
• Storyboard generation — Build multi-shot videos with separate prompts per segment.
• Advertising concepts — Prototype product, brand, or social video ideas.
• Cinematic scene generation — Create visually rich scenes with controlled aspect ratio.
• Creative iteration — Test variations of motion, camera direction, and scene style.

Pro Tips

• Use clear subject, action, camera, and style details in the prompt.
• Use multi_prompt when you need multiple shots instead of one continuous prompt.
• Make the multi_prompt shot durations sum exactly to the selected total duration.
• Use consistent wording across storyboard segments for better visual continuity.
• Choose the aspect ratio before writing composition-heavy prompts.

Notes

• Provide either prompt or multi_prompt.
• prompt and multi_prompt are mutually exclusive.
• duration is the total video duration for both single-prompt and multi_prompt generation.
• multi_prompt supports per-shot durations.
• For multi_prompt, the shot durations must sum exactly to the selected duration (3–15 seconds).

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'
{
  "aspect_ratio": "16:9",
  "duration": 5,
  "cfg_scale": 0.5
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/kwaivgi/kling-v3-turbo-pro/text-to-video" \
  -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