### Kling-O3 Image to Video API Examples

Source: https://docs.evolink.ai/en/api-manual/video-series/kling/kling-o3-image-to-video

Examples demonstrating different configurations for the kling-o3-image-to-video API. These snippets showcase minimal setup, multi-image inputs, and advanced parameter usage.

```json
{
  "model": "kling-o3-image-to-video",
  "prompt": "The person in the image slowly turns their head and smiles",
  "image_start": "https://example.com/portrait.jpg",
  "duration": 5,
  "quality": "720p"
}
```

```json
{
  "model": "kling-o3-image-to-video",
  "prompt": "Transition from daytime to nighttime",
  "image_start": "https://example.com/day.jpg",
  "image_end": "https://example.com/night.jpg",
  "duration": 10,
  "quality": "1080p",
  "sound": "on"
}
```

```json
{
  "model": "kling-o3-image-to-video",
  "prompt": "<<<element_1>>> dancing in the scene of <<<image_1>>>",
  "image_urls": [
    "https://example.com/scene.jpg"
  ],
  "duration": 8,
  "aspect_ratio": "16:9",
  "quality": "1080p",
  "model_params": {
    "element_list": [
      {
        "element_id": "123456"
      }
    ]
  }
}
```

```json
{
  "model": "kling-o3-image-to-video",
  "image_start": "https://example.com/portrait.jpg",
  "image_urls": [
    "https://example.com/bg.jpg"
  ],
  "duration": 10,
  "aspect_ratio": "16:9",
  "quality": "1080p",
  "model_params": {
    "multi_shot": true,
    "shot_type": "customize",
    "multi_prompt": [
      {
        "index": 1,
        "prompt": "<<<image_1>>> person sitting in a park",
        "duration": "5"
      },
      {
        "index": 2,
        "prompt": "Camera switches to <<<image_2>>> street view",
        "duration": "5"
      }
    ]
  }
}
```

--------------------------------

### Install OpenClaw Feishu Plugin

Source: https://docs.evolink.ai/en/integration-guide/openclaw-feishu

After installing the OpenClaw gateway, install the Feishu plugin using the OpenClaw CLI.

```bash
openclaw plugins install @openclaw/feishu
```

--------------------------------

### Install OpenCode via NPM (Windows)

Source: https://docs.evolink.ai/en/integration-guide/opencode

Install OpenCode on Windows using NPM after installing Node.js.

```bash
npm i -g opencode-ai@latest
```

--------------------------------

### Multiple images input example

Source: https://docs.evolink.ai/en/api-manual/language-series/gemini-3.1-flash-lite-preview/openai-sdk/openai-sdk-reference

This example demonstrates how to provide multiple images as input along with text.

```json
{
  "model": "gemini-3.1-flash-lite-preview",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Compare the differences between these two images"
        },
        {
          "type": "image_url",
          "image_url": {
            "url": "https://example.com/image1.png"
          }
        },
        {
          "type": "image_url",
          "image_url": {
            "url": "https://example.com/image2.png"
          }
        }
      ]
    }
  ]
}
```

--------------------------------

### Install Node.js using Homebrew (macOS)

Source: https://docs.evolink.ai/en/integration-guide/openclaw-telegram

Use Homebrew to install Node.js on macOS. Verify the installation afterwards.

```bash
brew install node

node --version
npm --version
```

--------------------------------

### Multimodal input example (text + image)

Source: https://docs.evolink.ai/en/api-manual/language-series/gemini-3.1-flash-lite-preview/openai-sdk/openai-sdk-reference

This example shows how to combine text and a single image for input.

```json
{
  "model": "gemini-3.1-flash-lite-preview",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": ">-
            Please describe the scene and main elements in this
            image in detail."
        },
        {
          "type": "image_url",
          "image_url": {
            "url": "https://example.com/image.png"
          }
        }
      ]
    }
  ]
}
```

--------------------------------

### Verify OpenCode Installation

Source: https://docs.evolink.ai/en/integration-guide/opencode

Run this command to check if OpenCode has been installed successfully and display its version.

```bash
opencode --version
```

--------------------------------

### Verify Git Installation (Windows)

Source: https://docs.evolink.ai/en/integration-guide/openclaw-auto

After installing Git on Windows, run this command in PowerShell to confirm the installation.

```bash
git --version
```

--------------------------------

### Verify Node.js and npm Installation (Windows)

Source: https://docs.evolink.ai/en/integration-guide/openclaw-auto

Run these commands in PowerShell after installing Node.js to verify the installation. It's recommended to run PowerShell as administrator.

```bash
node --version
npm --version
```

--------------------------------

### Install OpenClaw Gateway

Source: https://docs.evolink.ai/en/integration-guide/openclaw-telegram

Install the latest version of OpenClaw globally using npm.

```bash
npm install -g openclaw@latest
```

--------------------------------

### Verify Node.js and npm Installation

Source: https://docs.evolink.ai/en/integration-guide/gemini-cli

Check if Node.js and npm are installed and meet the version requirements (v20+).

```bash
node -v
npm -v
```

--------------------------------

### Run OpenClaw Onboarding

Source: https://docs.evolink.ai/en/integration-guide/openclaw-telegram

Execute the onboarding command to set up OpenClaw and install the background daemon service. Follow the on-screen prompts for installation mode.

```bash
openclaw onboard --install-daemon
```

--------------------------------

### Gemini-2.5-flash Quick Chat

Source: https://docs.evolink.ai/en/api-manual/language-series/gemini-2.5-flash/native-api/native-api-quickstart

Call the Gemini-2.5-flash model using the Google Native API format for synchronous processing and real-time conversation content. This example uses minimal parameters for a quick start.

```APIDOC
## POST /v1beta/models/gemini-2.5-flash:generateContent

### Description
Call Gemini-2.5-flash model using Google Native API format. Synchronous processing mode, returns conversation content in real-time. Minimal parameters for quick start. 💡 Need more features? Check [Full API Reference](./native-api-reference)

### Method
POST

### Endpoint
/v1beta/models/gemini-2.5-flash:generateContent

### Request Body
- **schema** (object) - Required - Defines the structure of the request payload.

### Responses
#### Success Response (200)
- **description**: Content generated successfully
- **content**: application/json
  - **schema**: Reference to `#/components/schemas/GenerateContentResponse`

#### Error Responses
- **400**: Invalid request parameters
- **401**: Unauthorized, invalid or expired token
- **402**: Insufficient quota, recharge required
- **403**: Access denied
- **404**: Resource not found
- **429**: Rate limit exceeded
- **500**: Internal server error
- **502**: Upstream service error
- **503**: Service temporarily unavailable
```

--------------------------------

### Create Chat Completion

Source: https://docs.evolink.ai/en/api-manual/language-series/gemini-2.5-flash-lite/openai-sdk/openai-sdk-quickstart

Call the Gemini 2.5 Flash Lite model using the OpenAI SDK format for synchronous chat completions. This example demonstrates a minimal setup for quick integration.

```APIDOC
## POST /v1/chat/completions

### Description
Call gemini-2.5-flash-lite model using OpenAI SDK format. Synchronous processing mode, returns conversation content in real-time. Minimal parameters for quick start.

### Method
POST

### Endpoint
/v1/chat/completions

### Request Body
- **schema** (object) - Required - The request body schema for chat completion.

### Responses
#### Success Response (200)
- **description** (string) - Chat completion generated successfully.

#### Error Responses
- **400** (object) - Invalid request parameters.
- **401** (object) - Unauthorized, invalid or expired token.
- **402** (object) - Insufficient quota, recharge required.
- **403** (object) - Access denied.
- **404** (object) - Resource not found.
- **429** (object) - Rate limit exceeded.
- **500** (object) - Internal server error.
- **502** (object) - Upstream service error.
- **503** (object) - Service temporarily unavailable.
```

--------------------------------

### Video Edit with Reference Image Request (Fast Mode)

Source: https://docs.evolink.ai/en/api-manual/video-series/kling/kling-o1-video-edit-fast

This example demonstrates how to perform a fast video edit using a reference image to guide the editing style. The task is processed asynchronously.

```json
{
  "model": "kling-o1-video-edit-fast",
  "prompt": "Make the video more cinematic",
  "video_urls": [
    "https://example.com/original-video.mp4"
  ],
  "image_urls": [
    "https://example.com/reference.jpg"
  ]
}
```

--------------------------------

### Create Qwen Voice Design (Full Parameters)

Source: https://docs.evolink.ai/en/api-manual/audio-series/qwen-tts/qwen-voice-design

This example demonstrates creating a custom voice profile using all available parameters, including language, sample rate, and response format, in addition to the core voice and preview details.

```json
{
  "model": "qwen-voice-design",
  "voice_prompt": ">-\n                    A young, energetic female voice with a fast pace and\n                    noticeable upward intonation, suitable for introducing\n                    fashion products",
  "preview_text": "Hi everyone, welcome to today's fashion review!",
  "preferred_name": "fashion_host",
  "language": "en",
  "sample_rate": 48000,
  "response_format": "mp3"
}
```

--------------------------------

### Gemini-3.1-pro Chat Completion

Source: https://docs.evolink.ai/en/api-manual/language-series/gemini-3.1-pro/openai-sdk/openai-sdk-quickstart

This endpoint allows you to interact with the Gemini-3.1-pro model using the OpenAI SDK format. It supports synchronous processing and returns conversation content in real-time. This is a minimal setup for a quick start.

```APIDOC
## POST /v1/chat/completions

### Description
Call Gemini-3.1-pro model using OpenAI SDK format. Synchronous processing mode, returns conversation content in real-time. Minimal parameters for quick start.

### Method
POST

### Endpoint
/v1/chat/completions

### Request Body
- **schema** (object) - Required - Defines the structure of the chat completion request.

### Responses
#### Success Response (200)
- **description**: Chat completion generated successfully
- **content**: application/json
  - **schema**: Reference to the ChatCompletionResponse schema.

#### Error Responses
- **400**: Invalid request parameters. Example: `{"error": {"code": 400, "message": "Invalid request parameters", "type": "invalid_request_error"}}`
- **401**: Unauthorized, invalid or expired token. Example: `{"error": {"code": 401, "message": "Invalid or expired token", "type": "authentication_error"}}`
- **402**: Insufficient quota, recharge required. Example: `{"error": {"code": 402, "message": "Insufficient quota", "type": "insufficient_quota_error", "fallback_suggestion": "https://evolink.ai/dashboard/billing"}}`
- **403**: Access denied. Example: `{"error": {"code": 403, "message": "Access denied for this model", "type": "permission_error", "param": "model"}}`
- **404**: Resource not found. Example: `{"error": {"code": 404, "message": "Specified model not found", "type": "not_found_error", "param": "model", "fallback_suggestion": "gemini-3.1-pro-preview"}}`
- **429**: Rate limit exceeded. Example: `{"error": {"code": 429, "message": "Rate limit exceeded", "type": "rate_limit_error", "fallback_suggestion": "retry after 60 seconds"}}`
- **500**: Internal server error. Example: `{"error": {"code": 500, "message": "Internal server error", "type": "internal_server_error", "fallback_suggestion": "try again later"}}`
- **502**: Upstream service error. Example: `{"error": {"code": 502, "message": "Upstream AI service unavailable", "type": "upstream_error", "fallback_suggestion": "try different model"}}`
- **503**: Service temporarily unavailable. Example: `{"error": {"code": 503, "message": "Service temporarily unavailable", "type": "service_unavailable_error"}}`
```

--------------------------------

### Kling-V3 Image to Video API Examples

Source: https://docs.evolink.ai/en/api-manual/video-series/kling/kling-v3-image-to-video

Examples demonstrating different use cases for the kling-v3-image-to-video model, including basic image-to-video, first and last frame transitions, element control, and multi-shot generation.

```json
{
  "model": "kling-v3-image-to-video",
  "prompt": "The person in the scene slowly turns their head and smiles",
  "image_start": "https://example.com/portrait.jpg",
  "duration": 5,
  "quality": "720p"
}
```

```json
{
  "model": "kling-v3-image-to-video",
  "prompt": "Transition from day to night",
  "image_start": "https://example.com/day.jpg",
  "image_end": "https://example.com/night.jpg",
  "duration": 10,
  "quality": "1080p",
  "sound": "on"
}
```

```json
{
  "model": "kling-v3-image-to-video",
  "prompt": "<<<element_1>>> is dancing in the scene",
  "image_start": "https://example.com/background.jpg",
  "duration": 8,
  "quality": "1080p",
  "model_params": {
    "element_list": [
      {
        "element_id": "123456"
      }
    ]
  }
}
```

```json
{
  "model": "kling-v3-image-to-video",
  "image_start": "https://example.com/portrait.jpg",
  "duration": 10,
  "quality": "1080p",
  "model_params": {
    "multi_shot": true,
    "shot_type": "customize",
    "multi_prompt": [
      {
        "index": 1,
        "prompt": "A person sitting in the park",
        "duration": "5"
      },
      {
        "index": 2,
        "prompt": "Camera switches to a full street view",
        "duration": "5"
      }
    ]
  }
}
```

--------------------------------

### Gemini-3.0-pro Chat Completion

Source: https://docs.evolink.ai/en/api-manual/language-series/gemini-3.0-pro/openai-sdk/openai-sdk-quickstart

This endpoint allows you to interact with the Gemini-3.0-pro model using the OpenAI SDK format. It supports synchronous processing and returns conversation content in real-time. This is a minimal setup for a quick start.

```APIDOC
## POST /v1/chat/completions

### Description
Call Gemini-3.0-pro model using OpenAI SDK format. Synchronous processing mode, returns conversation content in real-time. Minimal parameters for quick start.

### Method
POST

### Endpoint
/v1/chat/completions

### Request Body
- **schema** (object) - Required - The request body schema for chat completion.

### Responses
#### Success Response (200)
- **description** (string) - Chat completion generated successfully.
- **content** (object) - The response content.

#### Error Responses
- **400** - Invalid request parameters.
- **401** - Unauthorized, invalid or expired token.
- **402** - Insufficient quota, recharge required.
- **403** - Access denied.
- **404** - Resource not found.
- **429** - Rate limit exceeded.
- **500** - Internal server error.
- **502** - Upstream service error.
- **503** - Service temporarily unavailable.
```

--------------------------------

### Install OpenCode via Scoop (Windows)

Source: https://docs.evolink.ai/en/integration-guide/opencode

Install OpenCode on Windows using the Scoop package manager. Ensure the 'extras' bucket is added first.

```bash
scoop bucket add extras
scoop install extras/opencode
```

--------------------------------

### Correct Prompt Formatting Example

Source: https://docs.evolink.ai/en/api-manual/image-series/midjourney/midjourney-v7-prompt-guide

Demonstrates the correct way to format a prompt with parameters, ensuring a space before the double dash and no punctuation within parameters.

```text
a yellow Persian cat --ar 1:2 --s 500
```

--------------------------------

### Gemini-2.5-pro Quick Chat

Source: https://docs.evolink.ai/en/api-manual/language-series/gemini-2.5-pro/native-api/native-api-quickstart

Call the Gemini-2.5-pro model using the Google Native API format for synchronous processing and real-time conversation content. This example uses minimal parameters for a quick start. For more features, refer to the Full API Reference.

```APIDOC
## POST /v1beta/models/gemini-2.5-pro:generateContent

### Description
This endpoint allows you to interact with the Gemini-2.5-pro model for content generation in a synchronous mode. It's designed for a quick start with essential parameters.

### Method
POST

### Endpoint
https://direct.evolink.ai/v1beta/models/gemini-2.5-pro:generateContent

### Request Body
- **Required**: true
  - **Content**: application/json
    - **Schema**: Reference to `#/components/schemas/GenerateContentQuickRequest`

### Responses
#### Success Response (200)
- **Description**: Content generated successfully
- **Content**: application/json
  - **Schema**: Reference to `#/components/schemas/GenerateContentResponse`

#### Error Responses
- **400 Bad Request**: Invalid request parameters. Example response includes `error.code: 400`, `error.message: Invalid request parameters`, `error.type: invalid_request_error`.
- **401 Unauthorized**: Invalid or expired token. Example response includes `error.code: 401`, `error.message: Invalid or expired token`, `error.type: authentication_error`.
- **402 Insufficient Quota**: Recharge required. Example response includes `error.code: 402`, `error.message: Insufficient quota`, `error.type: insufficient_quota_error`, `fallback_suggestion: https://evolink.ai/dashboard/billing`.
- **403 Forbidden**: Access denied. Example response includes `error.code: 403`, `error.message: Access denied for this model`, `error.type: permission_error`.
- **404 Not Found**: Resource not found. Example response includes `error.code: 404`, `error.message: Model not found`, `error.type: not_found_error`.
- **429 Too Many Requests**: Rate limit exceeded. Example response includes `error.code: 429`, `error.message: Rate limit exceeded`, `error.type: rate_limit_error`, `fallback_suggestion: retry after 60 seconds`.
- **500 Internal Server Error**: Server error. Example response includes `error.code: 500`, `error.message: Internal server error`, `error.type: internal_server_error`, `fallback_suggestion: try again later`.
- **502 Bad Gateway**: Upstream service error. Example response includes `error.code: 502`, `error.message: Upstream AI service unavailable`, `error.type: upstream_error`, `fallback_suggestion: try again later`.
- **503 Service Unavailable**: Service temporarily unavailable. Example response includes `error.code: 503`.
```

--------------------------------

### Kling-O1 Video Edit API Request with Reference Image

Source: https://docs.evolink.ai/en/api-manual/video-series/kling/kling-o1-video-edit

This example shows how to perform video editing with a reference image. In addition to the video URL and prompt, it includes a URL for a reference image to guide the editing process. Generated video links are valid for 24 hours.

```json
{
  "model": "kling-o1-video-edit",
  "prompt": "Make the video more cinematic",
  "video_urls": [
    "https://example.com/original-video.mp4"
  ],
  "image_urls": [
    "https://example.com/reference.jpg"
  ]
}
```

--------------------------------

### Create Qwen Voice Design (Minimal)

Source: https://docs.evolink.ai/en/api-manual/audio-series/qwen-tts/qwen-voice-design

Use this example to create a custom voice profile with minimal required parameters. It specifies the model, a voice prompt, and preview text, with an optional preferred name.

```json
{
  "model": "qwen-voice-design",
  "voice_prompt": ">-\n                    A calm middle-aged male news anchor with a deep, resonant\n                    voice, rich in magnetism, steady pace, and clear\n                    articulation",
  "preview_text": ">-\n                    Good evening, listeners. Welcome to the evening news\n                    broadcast.",
  "preferred_name": "announcer"
}
```

--------------------------------

### Basic Text to Video Generation

Source: https://docs.evolink.ai/en/api-manual/video-series/kling/kling-o3-text-to-video

Use this example for simple text-to-video generation. Specify the prompt, desired duration, aspect ratio, and quality.

```json
{
  "model": "kling-o3-text-to-video",
  "prompt": "A cat running on a grassy field under bright sunshine",
  "duration": 5,
  "aspect_ratio": "16:9",
  "quality": "720p"
}
```

--------------------------------

### Text to Video Example Request

Source: https://docs.evolink.ai/en/api-manual/video-series/grok/grok-imagine-text-to-video

An example request body for generating a video from a text prompt using the Grok Imagine model. Specifies the model, prompt, duration, mode, aspect ratio, and quality.

```json
{
  "model": "grok-imagine-text-to-video-beta",
  "prompt": "A cat playing piano in a jazz club",
  "duration": 10,
  "mode": "fun",
  "aspect_ratio": "16:9",
  "quality": "720p"
}
```

--------------------------------

### System prompt example

Source: https://docs.evolink.ai/en/api-manual/language-series/gemini-3.1-flash-lite-preview/openai-sdk/openai-sdk-reference

Customize the AI's behavior and role by providing a system prompt.

```json
{
  "model": "gemini-3.1-flash-lite-preview",
  "messages": [
    {
      "role": "system",
      "content": ">-
        You are a professional Python programming assistant,
        answering questions concisely."
    },
    {
      "role": "user",
      "content": "How to read a file?"
    }
  ]
}
```

--------------------------------

### Install OpenClaw Gateway Globally

Source: https://docs.evolink.ai/en/integration-guide/openclaw

Install the latest version of OpenClaw Gateway globally using npm. Verify the installation by checking the version.

```bash
npm install -g openclaw@latest

openclaw --version
```

--------------------------------

### Initialize EvoLink Provider in OpenCode

Source: https://docs.evolink.ai/en/integration-guide/opencode

Configure OpenCode to use EvoLink API by running the login command and following the interactive prompts to select 'other' provider, enter 'evolink-anthropic' as the Provider ID, and any value for the API Key token.

```bash
opencode auth login
```

--------------------------------

### Verify Codex CLI Installation

Source: https://docs.evolink.ai/en/integration-guide/codex-cli

Confirm that the Codex CLI has been installed successfully by checking its version. A successful installation will display the version number.

```bash
codex --version
```

--------------------------------

### Install Git using Xcode Command Line Tools (macOS)

Source: https://docs.evolink.ai/en/integration-guide/openclaw-auto

An alternative method to install Git on macOS by using the Xcode Command Line Tools.

```bash
xcode-select --install
```

--------------------------------

### Verify Claude Code CLI Installation

Source: https://docs.evolink.ai/en/integration-guide/claude-code-cli

Confirm that the Claude Code CLI has been installed successfully by checking its version. A successful installation will display the version number.

```bash
claude --version
```

--------------------------------

### VideoRetalk Generation with Multi-Face and Extension

Source: https://docs.evolink.ai/en/api-manual/video-series/videoretalk/videoretalk-video-generate

This example demonstrates generating a lip-sync video for a scene with multiple faces and enables video extension. The `ref_image_url` is crucial for specifying the target person when multiple faces are present.

```json
{
  "model": "videoretalk",
  "video_url": "https://example.com/two-people-talking.mp4",
  "audio_url": "https://example.com/long-speech.wav",
  "ref_image_url": "https://example.com/target-person-face.jpg",
  "video_extension": true,
  "query_face_threshold": 150
}
```

--------------------------------

### Install Claude Code CLI on macOS/Linux

Source: https://docs.evolink.ai/en/integration-guide/claude-code-cli

Use this command to install the Claude Code CLI on macOS or Linux systems. Ensure you have curl installed. If you encounter permission errors, prepend the command with 'sudo'.

```bash
curl -fsSL https://claude.ai/install.sh | bash
```

--------------------------------

### Suno Persona Creation with Vocal Extraction Window

Source: https://docs.evolink.ai/en/api-manual/audio-series/suno/suno-persona-creation

This example demonstrates how to create a persona while specifying a vocal extraction window using `vocal_start` and `vocal_end` parameters. The difference between `vocal_end` and `vocal_start` must be between 10 and 30 seconds.

```json
{
  "model": "suno-persona",
  "model_params": {
    "source_task_id": "task-unified-1774169216-ocqaqde7",
    "result_id": "4fcc4507-a7ae-4441-ad8a-465c2f61d5bb",
    "name": "Jazz Vocalist",
    "description": "Smooth jazz vocal style with warm tones and improvisational flair suitable for lounge music",
    "vocal_start": 15,
    "vocal_end": 35,
    "style": "jazz"
  }
}
```

--------------------------------

### Verify Gemini CLI Installation

Source: https://docs.evolink.ai/en/integration-guide/gemini-cli

Confirm that the Gemini CLI has been installed successfully by checking its version.

```bash
gemini --version
```