--- title: Kuaishou Technology | Scenario Docs --- > This page is auto-generated from model configurations. Last updated: 2026-10-09. This reference lists all available **Kuaishou Technology** image generation models and their parameters. Use these parameter names when calling the [Generation API](/api/postgeneratecustom/index.md). - [Kling O3 Image to Image](#kling-o3-image-to-image) - [Kling O3 Text to Image](#kling-o3-text-to-image) - [Kling V3 Image to Image](#kling-v3-image-to-image) - [Kling V3 Text to Image](#kling-v3-text-to-image) --- ## Kling O3 Image to Image Kling O3 Omni image-to-image - edit or combine up to 10 reference images into up to 9 images at 1K, 2K, or 4K, with optional face-controlled elements. **Model ID:** `model_kling-o3-i2i` **Capabilities:** `img2img` **LLM Markdown:** | Parameter | Type | Required | Default | Min | Max | Allowed Values | Description | | ----------------- | ------------ | -------- | ------- | --- | --- | ----------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `prompt` | string | Yes | - | - | - | - | Describe the edit or image to generate. Reference images as @Image1, @Image2, etc. (or @Image with one image) and elements as @Element1, @Element2, etc. Max 2500 characters. | | `referenceImages` | file_array | Yes | - | - | - | - | Reference images (1-10), referenced in the prompt as @Image1, @Image2, etc. Min 300x300 px, aspect ratio 0.4-2.5, max 10 MB each. | | `aspectRatio` | string | No | `auto` | - | - | `auto`, `16:9`, `9:16`, `1:1`, `4:3`, `3:4`, `3:2`, `2:3`, `21:9` | Aspect ratio of the generated images. Auto picks it from the input content. | | `resolution` | string | No | `1K` | - | - | `1K`, `2K`, `4K` | Output resolution. 4K costs twice as much per image. | | `numOutputs` | number | No | `1` | 1 | 9 | - | How many images to generate (1-9). Each image is billed. | | `elements` | inputs_array | No | - | - | - | - | Optional characters or objects to keep consistent (face control). Reference them in the prompt as @Element1, @Element2, etc. | ## Kling O3 Text to Image Kling O3 Omni text-to-image - up to 9 images per run at 1K, 2K, or 4K, with optional face-controlled elements. **Model ID:** `model_kling-o3-t2i` **Capabilities:** `txt2img` **LLM Markdown:** | Parameter | Type | Required | Default | Min | Max | Allowed Values | Description | | ------------- | ------------ | -------- | ------- | --- | --- | --------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------- | | `prompt` | string | Yes | - | - | - | - | Describe the image to generate. Reference elements as @Element1, @Element2, etc. Max 2500 characters. | | `aspectRatio` | string | No | `16:9` | - | - | `16:9`, `9:16`, `1:1`, `4:3`, `3:4`, `3:2`, `2:3`, `21:9` | Aspect ratio of the generated images. | | `resolution` | string | No | `1K` | - | - | `1K`, `2K`, `4K` | Output resolution. 4K costs twice as much per image. | | `numOutputs` | number | No | `1` | 1 | 9 | - | How many images to generate (1-9). Each image is billed. | | `elements` | inputs_array | No | - | - | - | - | Optional characters or objects to keep consistent (face control). Reference them in the prompt as @Element1, @Element2, etc. | ## Kling V3 Image to Image Kling V3 image-to-image - generate up to 9 images from a reference image at 1K or 2K, with optional face-controlled elements. **Model ID:** `model_kling-v3-i2i` **Capabilities:** `img2img` **LLM Markdown:** | Parameter | Type | Required | Default | Min | Max | Allowed Values | Description | | ------------- | ------------ | -------- | ------- | --- | --- | --------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------- | | `prompt` | string | Yes | - | - | - | - | Describe the image to generate from the reference. Reference elements as @Element1, @Element2, etc. Max 2500 characters. | | `image` | file | Yes | - | - | - | - | Reference image. Min 300x300 px, aspect ratio 0.4-2.5, max 10 MB. | | `aspectRatio` | string | No | `16:9` | - | - | `16:9`, `9:16`, `1:1`, `4:3`, `3:4`, `3:2`, `2:3`, `21:9` | Aspect ratio of the generated images. | | `resolution` | string | No | `1K` | - | - | `1K`, `2K` | Output resolution. 2K is high-res at the same price. | | `numOutputs` | number | No | `1` | 1 | 9 | - | How many images to generate (1-9). Each image is billed. | | `elements` | inputs_array | No | - | - | - | - | Optional characters or objects to keep consistent (face control). Reference them in the prompt as @Element1, @Element2, etc. | ## Kling V3 Text to Image Kling V3 text-to-image - up to 9 images per run at 1K or 2K, with optional face-controlled elements. **Model ID:** `model_kling-v3-t2i` **Capabilities:** `txt2img` **LLM Markdown:** | Parameter | Type | Required | Default | Min | Max | Allowed Values | Description | | ---------------- | ------------ | -------- | ------- | --- | --- | --------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------- | | `prompt` | string | Yes | - | - | - | - | Describe the image to generate. Reference elements as @Element1, @Element2, etc. Max 2500 characters. | | `negativePrompt` | string | No | - | - | - | - | Content to avoid. Kling recommends phrasing exclusions as negative sentences in the main prompt instead. | | `aspectRatio` | string | No | `16:9` | - | - | `16:9`, `9:16`, `1:1`, `4:3`, `3:4`, `3:2`, `2:3`, `21:9` | Aspect ratio of the generated images. | | `resolution` | string | No | `1K` | - | - | `1K`, `2K` | Output resolution. 2K is high-res at the same price. | | `numOutputs` | number | No | `1` | 1 | 9 | - | How many images to generate (1-9). Each image is billed. | | `elements` | inputs_array | No | - | - | - | - | Optional characters or objects to keep consistent (face control). Reference them in the prompt as @Element1, @Element2, etc. |