Kling V3 Text-to-Video API

Overview

Kling V3 is Kuaishou's third-generation video generation model in its text-to-video-only interface. It takes a text prompt as the sole input and switches between Standard (720p), Pro (1080p), and 4K resolution tiers via the mode field, supporting 4–15 second short video generation.

Compared to the same-series Kling V3 Omni, V3 has a simpler API structure—the request body is flat (prompt / mode / duration), with no image lists, reference videos, multi-shot editing, or audio generation. It is suited for scenarios requiring fast, low-cost text-to-video generation. On iCreat, it is invoked through an asynchronous task queue at kwaivgi/kling-v3/text-to-video.

kwaivgi/kling-v3/text-to-video

Core Capabilities

Pure text-driven generation. Generate video from a single prompt—no reference images or video assets needed. The prompt supports multi-line text (\n), with Chinese up to 2,000 characters and English up to 2,000 words.

Three resolution tiers. Switch via the mode field: std (720p) for the lowest cost, pro (1080p) for enhanced detail, and 4k (2160p) for premium delivery.

4–15 second duration. Supports video generation from a minimum of 4 seconds to a maximum of 15 seconds, billed per second—ideal for short videos, social media assets, and rapid prototyping.

Chinese and English prompts. Officially supports Chinese and English prompts, with strong comprehension of scene descriptions and action instructions in both languages.

Flat request structure. Unlike Wan 2.7 / HappyHorse 1.1 / Seedance 2.5's input + parameters nesting, Kling V3's request body is flat—prompt, mode, and duration sit at the JSON root level.

Model Comparison

Kling V3 vs Kling V3 Omni vs Wan 2.7 Text-to-Video

Input

Kling V3 text-to-video accepts only a text prompt—no images, videos, or character references.

Parameters

Mode Mapping

Note: Kling V3 uses std (abbreviated), while the same-series Kling V3 Omni uses standard (full word). Both share the pro and 4k values.

Pricing

Base prices below exclude reference video (this interface is text-to-video only—reference video is not supported).

Total Cost = Unit Price × Output Video Duration

Quick Start

1. Authentication

Get your API Key from https://icreat.ai/hub/keys.

2. Submit a Request

curl --fail-with-body --connect-timeout 10 --max-time 60 \
  -X POST "https://api.icreat.ai/v1/task/submit/kwaivgi/kling-v3/text-to-video" \
  -H "Authorization: Bearer ${ICREAT_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "示例多行\n文本",
    "mode": "std",
    "duration": 4
  }'

Response example:

{
  "task_id": "task-xxx"
}

3. Poll Task Status

curl --connect-timeout 10 --max-time 60 \
  -X POST "https://api.icreat.ai/v1/task/query-status" \
  -H "Authorization: Bearer ${ICREAT_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "task_id": "task-xxx"
  }'

When the status is SUCCEEDED, you can retrieve the result.

4. Get the Result

curl --connect-timeout 10 --max-time 60 \
  -X POST "https://api.icreat.ai/v1/task/get-result" \
  -H "Authorization: Bearer ${ICREAT_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "task_id": "task-xxx"
  }'

The server will return the task output.

Output

submit → task_id → query-status → get-result

After submitting, you receive a task_id. Poll query-status until the status becomes SUCCEEDED, then call get-result with the same task_id to retrieve the generated video.

Input Examples

Generate a 720p Video

{
  "prompt": "一只小狗在草地上奔跑",
  "mode": "std",
  "duration": 5
}

Generate a 1080p Video

{
  "prompt": "一只小狗在草地上奔跑",
  "mode": "pro",
  "duration": 5
}

Generate a 4K Video

{
  "prompt": "一只小狗在草地上奔跑",
  "mode": "4k",
  "duration": 5
}

Use Cases

Rapid prototyping and concept validation. Use the lowest-cost std mode to quickly validate prompt effects, composition, and motion direction, then switch to pro or 4k for the final high-quality cut.

Short video and social media. The 4–15 second duration covers short-video needs on Douyin, Kuaishou, Xiaohongshu, and similar platforms, with text-only input reducing asset preparation overhead.

Advertising and marketing assets. pro mode generates 1080p ad creative clips; 4k mode is for brand films and premium delivery.

Batch variant generation. The flat request structure makes it easy to programmatically construct prompts and generate multiple variants for A/B testing.

Educational and demo videos. Describe scenes and action flows in natural language to generate teaching demos and concept animations.

Limitations

Maximum duration is 15 seconds—long videos cannot be generated via this interface (use Kling V3 Omni for up to 120 seconds when needed).

No support for image, video, or character references—all visual content is driven by the text prompt.

No support for audio generation, negative prompts, aspect ratio selection, or watermark control (use Kling V3 Omni or Wan 2.7 when these features are needed).

The mode value is std (not standard), differing from Kling V3 Omni's standard—be careful when migrating code.

Non-Chinese/English prompts are untested and may reduce generation quality.

The same prompt produces different variants on each run (by design); use more detailed prompts to constrain the model when deterministic output is needed.

FAQ

What's the difference between Kling V3 and Kling V3 Omni? Which should I choose?

V3 is a text-to-video-only interface that accepts only text prompts, up to 15 seconds, with a simple API structure. V3 Omni is the full-featured version supporting image/video/character references, multi-shot editing, audio generation, and up to 120 seconds of output. Choose V3 for quick text-to-short-video; choose V3 Omni for multimodal references, long videos, or complex narratives. Both share identical 720P / 1080P / 4K base pricing.

Are std and standard the same mode?

No. Kling V3 uses std, while Kling V3 Omni uses standard—both correspond to 720p output, but the string values differ. You must update the mode value when migrating code.

Why is V3's pricing the same as V3 Omni's?

Both share the same underlying model and billing baseline. V3 Omni's extra costs come from optional features (reference video +50%, multi-shot, etc.); under pure text-to-video, both are priced identically.

Is 4K mode worth it?

4K pricing is $0.42/s—5× the `std` rate. A 15-second 4K video costs $6.30. Use 4K only when the final deliverable requires that resolution; use std for testing and preview.

Are failed generations billed?

No. Failed tasks (FAILED) incur no cost and can be safely resubmitted. Any regeneration after SUCCEEDED is a new request and will be billed again.

How long can the prompt be?

Chinese up to 2,000 characters, English up to 2,000 words. Multi-line prompts with \n are supported for organizing scene descriptions and action instructions.