UniAll AI SEO/GEO · Kling V3 · 2026-05-31

Kling V3 API overview

Kling V3 is a video generation model available on UniAll AI for teams building short-form video, product clips, ad creatives, storyboard shots, and automated video workflows. The public model id for the standard silent variant is `kling-v3-std-silent`.

The model supports three main generation modes:

• Text to video: generate a video from a prompt.
• Image to video: animate a reference image with prompt guidance.
• First/last frame video: provide a starting frame and ending frame to guide the motion path.

Requests are asynchronous and use the `/v1/videos/generations` endpoint with `POST`. This makes Kling V3 suitable for backend pipelines where your application submits a task, polls or handles completion, and stores the generated video result.

Pricing model

Kling V3 is billed per generated second. On UniAll AI, the standard silent variant starts at `$0.05712` per second, shown in the user price profile as about `¥0.41 / second`. Final cost depends on the selected variant, duration, and whether audio or higher quality output is enabled.

Available user-facing variant prices include:

| Variant | Model ID | Price | |---|---:|---:| | Standard Silent | `kling-v3-std-silent` | `$0.05712 / second` | | Standard Audio | `kling-v3-std-audio` | `$0.08568 / second` | | Pro Silent | `kling-v3-pro-silent` | `$0.07616 / second` | | Pro Audio | `kling-v3-pro-audio` | `$0.11424 / second` | | 4K Silent | `kling-v3-4k-silent` | `$0.2856 / second` | | 4K Audio | `kling-v3-4k-audio` | `$0.2856 / second` |

For example, a 5-second generation on `kling-v3-std-silent` is estimated at `5 × $0.05712 = $0.2856`, before any account-specific billing rules. Failed generations are eligible for refund according to the platform configuration.

Supported parameters

Kling V3 supports output durations from 3 to 15 seconds. The default duration is 5 seconds. Aspect ratio options include `16:9`, `9:16`, and `1:1`, which cover landscape video, vertical short-form formats, and square social assets.

Resolution options include `standard`, `pro`, and `4k`. Each request produces one video per generation. Image inputs can be PNG, JPEG, or WebP, depending on the mode.

Example API request

```json { "model": "kling-v3-std-silent", "generation_mode": "image_to_video", "prompt": "A cinematic product reveal, soft studio lighting, smooth camera movement.", "image_url": "https://example.com/reference.png", "duration": 5, "aspect_ratio": "16:9", "resolution": "standard", "video_count": 1 } ```

For text-to-video, use `generation_mode: "text_to_video"` and provide a prompt. For first/last-frame generation, use `generation_mode: "first_last_frame"` with both `first_image_url` and `last_image_url`.

Who should use Kling V3 API

Kling V3 is useful for developers, agencies, ecommerce teams, and AI product builders who need programmable video generation rather than manual one-off creation. It fits workflows such as ad concept testing, product demo generation, vertical video production, storyboard iteration, and creative automation.

Teams comparing pricing should look beyond the base per-second rate. The practical cost depends on duration, resolution tier, audio requirements, retry behavior, and how many generations are needed to reach an approved result.