1. Veo
TikHub-AI-Proxy
  • Overview (PLEASE READ)
  • Streaming API
  • OpenAI
    • OpenAI response
      POST
    • OpenAI embeddings
      POST
    • OpenAI audio transcription
      POST
    • OpenAI chat completion
      POST
  • Claude
    • Claude chat completion
      POST
    • Claude message
      POST
  • DeepSeek
    • DeepSeek chat completion
      POST
  • Sora
    • Sora video generation
      POST
    • Get Sora video status
      GET
    • Download Sora video content
      GET
    • Remix Sora video
      POST
    • List Sora videos
      GET
    • Delete Sora video
      DELETE
  • Gemini
    • Gemini content
      POST
  • Seedance
    • Seedance video generation
      POST
    • Retrieve Seedance task
      GET
  • Kling
    • Kling text-to-video
      POST
    • Retrieve Kling text-to-video task
      GET
    • Kling image-to-video
      POST
    • Retrieve Kling image-to-video task
      GET
  • Veo
    • Veo video generation
      POST
    • Fetch Veo video generation status
      POST
  • Schemas
    • ChatCompletionRequest
    • ChatMessage
    • Tool
    • ToolCall
    • ChatCompletionResponse
    • ChatCompletionChoice
    • ContentFilterResults
    • UsageInfo
    • EmbeddingRequest
    • EmbeddingResponse
    • TranscriptionRequest
    • TranscriptionResponse
    • ClaudeMessageRequest
    • ClaudeMessageResponse
    • VideoCreateRequest
    • VideoResponse
    • VideoRemixRequest
    • GeminiGenerateContentRequest
    • VideoListResponse
    • GeminiContent
    • VideoDeleteResponse
    • GeminiGenerationConfig
    • ResponseRequest
    • GeminiGenerateContentResponse
    • ResponseObject
    • SeedanceTaskRequest
    • SeedanceTaskCreateResponse
    • SeedanceTaskResponse
    • KlingText2VideoRequest
    • KlingImage2VideoRequest
    • KlingTaskResponse
    • KlingTaskDetailResponse
    • KlingTaskListResponse
    • VeoGenerateRequest
    • VeoOperationResponse
    • VeoFetchOperationRequest
    • VeoFetchOperationResponse
  1. Veo

Veo video generation

POST
https://ai.tikhub.io/vertex/v1/publishers/google/models/{model}:predictLongRunning
Veo
Creates a video generation task using Google Veo models. The task is asynchronous - use the fetchPredictOperation endpoint to poll for status until done is true.

How To Use#

Specify the model in the path and provide a text prompt in the request payload.
Tip: If you receive an unsupported action error, this is typically a transient issue. Please retry the request.

Supported Models#

veo-2.0-generate-001: Veo 2.0 - base video generation (no audio, no resolution setting)
veo-3.0-generate-001: Veo 3.0 - supports audio generation, resolution up to 1080p
veo-3.0-fast-generate-001: Veo 3.0 Fast - faster generation with audio, resolution up to 1080p
veo-3.1-generate-001: Veo 3.1 - latest model, supports audio, resolution up to 4k, reference images, and video extension
veo-3.1-fast-generate-001: Veo 3.1 Fast - faster generation with all 3.1 features

Generation Modes#

Text-to-Video: Provide only a prompt in instances
Image-to-Video: Provide prompt + image object (all models)
Image Interpolation: Provide prompt + image + lastFrame (all models)
Reference Images: Provide prompt + referenceImages array with up to 3 style/content references (Veo 3.1 only)
Video Extension: Provide prompt + video object (Veo 3.1 only, requires durationSeconds: 8 and resolution: 720p)

Important Notes#

image + lastFrame, referenceImages, and video are mutually exclusive
1080p and 4k resolution require durationSeconds: 8
referenceImages requires durationSeconds: 8
Veo 2 duration options: 5, 6, 8 seconds. Veo 3/3.1 duration options: 4, 6, 8 seconds

Task Workflow#

Video generation is asynchronous. After submitting a request, you receive an operation name. Use the fetch endpoint to poll until the operation is done.

Request

Authorization
Bearer Token
Provide your bearer token in the
Authorization
header when making requests to protected resources.
Example:
Authorization: Bearer ********************
or
Path Params

Body Params application/jsonRequired

Examples
{
    "instances": [
        {
            "prompt": "A fast-tracking shot through a bustling dystopian sprawl with bright neon signs, flying cars and mist, night, lens flare, volumetric lighting"
        }
    ],
    "parameters": {
        "aspectRatio": "16:9",
        "durationSeconds": 8,
        "generateAudio": true,
        "resolution": "720p"
    }
}

Request Code Samples

Shell
JavaScript
Java
Swift
Go
PHP
Python
HTTP
C
C#
Objective-C
Ruby
OCaml
Dart
R
Request Request Example
Shell
JavaScript
Java
Swift
cURL
curl --location --request POST 'https://ai.tikhub.io/vertex/v1/publishers/google/models/veo-3.0-generate-001:predictLongRunning' \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data-raw '{
    "instances": [
        {
            "prompt": "A fast-tracking shot through a bustling dystopian sprawl with bright neon signs, flying cars and mist, night, lens flare, volumetric lighting"
        }
    ],
    "parameters": {
        "aspectRatio": "16:9",
        "durationSeconds": 8,
        "generateAudio": true,
        "resolution": "720p"
    }
}'

Responses

🟢200OK
application/json
Video generation task submitted successfully
Body

Example
{
    "name": "models/veo-3.0-generate-001/operations/6a5663c1-43b1-42eb-a083-b32ed06e3180"
}
🟠400Bad Request
🟠401Unauthorized
🔴500Internal Server Error
Modified at 2026-03-02 20:33:07
Previous
Retrieve Kling image-to-video task
Next
Fetch Veo video generation status
Built with