Veo 3.1 Video Generation Models Added

Date: 2025-11-18 / (1404-08-27)

Summary

We announce support for Google's Veo 3.1 video generation models, offering developers two options for creating AI-generated videos: veo-3.1-generate-preview for high-quality output and veo-3.1-fast-generate-preview optimized for speed. Both models are accessible through the v1/videos endpoint with support for text-to-video and image-to-video generation.

---

Details

Google Veo 3.1 Video Generation

We introduce Google's Veo 3.1 series, the latest advancement in AI video generation technology. These models enable developers to create high-quality videos from text prompts or reference images, with improved audio generation and enhanced visual quality.

Available Models

• veo-3.1-generate-preview: Delivers superior output quality with rich native audio, natural conversations, and synchronized sound effects. Best for production-ready content requiring maximum visual and audio fidelity.

• veo-3.1-fast-generate-preview: Speed-optimized version that maintains high quality while providing faster generation times. Ideal for rapid iteration, high-volume projects, and applications requiring quick turnaround.

Key Features:

• Native Audio Generation: Videos include synchronized audio with natural sound effects and ambient noise
• Enhanced Image-to-Video: Improved prompt adherence and character consistency across scenes
• Flexible Duration: Generate videos of 4, 6, or 8 seconds (default: 8 seconds)
• Multiple Resolutions: Support for 720p and 1080p output (16:9 aspect ratio)
• Aspect Ratio Options: 16:9 (landscape) and 9:16 (portrait) formats
• Reference Images: Guide generation with up to 3 reference images for character/style consistency
• Asynchronous Processing: Poll for completion
• Video Extension: Extend existing videos to create longer sequences

Pricing Details:

Model	Cost Per Second
veo-3.1-fast-generate-preview	$0.15/second
veo-3.1-generate-preview	$0.40/second

API Request/Response Examples

Generate a Video

Bash

curl https://api.avalai.ir/v1/videos \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "veo-3.1-fast-generate-preview",
    "prompt": "A cat playing with a ball of yarn in a sunny garden",
    "seconds": "4"
  }'

Python

from openai import OpenAI

client = OpenAI(api_key="your-avalai-api-key", base_url="https://api.avalai.ir/v1")

video = client.videos.create(
    model="veo-3.1-fast-generate-preview",
    prompt="A cat playing with a ball of yarn in a sunny garden",
    seconds="4",
)

print(f"Video ID: {video.id}")

JavaScript

import { OpenAI } from "openai";

const client = new OpenAI({
  apiKey: process.env.AVALAI_API_KEY,
  baseURL: "https://api.avalai.ir/v1",
});

const video = await client.videos.create({
  model: "veo-3.1-fast-generate-preview",
  prompt: "A cat playing with a ball of yarn in a sunny garden",
  seconds: "4",
});

console.log(`Video ID: ${video.id}`);

Response

{
  "id": "video_abc123def456",
  "object": "video",
  "created_at": 1731916800,
  "status": "processing",
  "model": "veo-3.1-fast-generate-preview",
  "progress": 0,
  "seconds": "4"
}

Check Video Status

Bash

curl https://api.avalai.ir/v1/videos/video_abc123def456 \
  -H "Authorization: Bearer $AVALAI_API_KEY"

Python

import time

# Poll for completion
while True:
    video_status = client.videos.retrieve("video_abc123def456")

    if video_status.status == "completed":
        print(f"Video ready! ID: {video_status.id}")
        break
    elif video_status.status == "failed":
        print(f"Generation failed: {video_status.error}")
        break

    print(f"Status: {video_status.status}, Progress: {video_status.progress}%")
    time.sleep(10)

JavaScript

// Poll for completion
while (true) {
  const videoStatus = await client.videos.retrieve("video_abc123def456");
  
  if (videoStatus.status === "completed") {
    console.log(`Video ready! ID: ${videoStatus.id}`);
    break;
  } else if (videoStatus.status === "failed") {
    console.log(`Generation failed: ${videoStatus.error}`);
    break;
  }
  
  console.log(`Status: ${videoStatus.status}, Progress: ${videoStatus.progress}%`);
  await new Promise(resolve => setTimeout(resolve, 10000));
}

Response (Completed)

{
  "id": "video_abc123def456",
  "object": "video",
  "created_at": 1731916800,
  "status": "completed",
  "model": "veo-3.1-fast-generate-preview",
  "progress": 100,
  "seconds": "4"
}

Download Video Content

Bash

curl https://api.avalai.ir/v1/videos/video_abc123def456/content \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  --output video.mp4

Python

content = client.videos.download_content("video_abc123def456")
with open("video.mp4", "wb") as f:
    f.write(content.read())
print("Video downloaded successfully!")

JavaScript

const content = await client.videos.downloadContent("video_abc123def456");
const buffer = Buffer.from(await content.arrayBuffer());
require('fs').writeFileSync('video.mp4', buffer);
console.log("Video downloaded successfully!");

Advanced Features

Image-to-Video Generation

Use reference images to guide video generation and ensure visual consistency:

Bash

curl https://api.avalai.ir/v1/videos \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -F "model=veo-3.1-generate-preview" \
  -F "prompt=The landscape comes alive with flowing water and moving clouds" \
  -F "input_reference=@reference_image.jpg" \
  -F "seconds=8"

Python

video = client.videos.create(
    model="veo-3.1-generate-preview",
    prompt="The landscape comes alive with flowing water and moving clouds",
    input_reference=open("reference_image.jpg", "rb"),
    seconds="8",
)

print(f"Image-to-video generation started: {video.id}")

JavaScript

import fs from 'fs';

const video = await client.videos.create({
  model: "veo-3.1-generate-preview",
  prompt: "The landscape comes alive with flowing water and moving clouds",
  input_reference: fs.createReadStream("reference_image.jpg"),
  seconds: "8",
});

console.log(`Image-to-video generation started: ${video.id}`);

Control Aspect Ratio and Resolution

Specify video dimensions using the size parameter:

Bash

# Landscape 16:9 at 1080p
curl https://api.avalai.ir/v1/videos \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "veo-3.1-generate-preview",
    "prompt": "Aerial drone shot of a coastal city at sunset",
    "size": "1920x1080",
    "seconds": "8"
  }'

# Portrait 9:16
curl https://api.avalai.ir/v1/videos \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "veo-3.1-fast-generate-preview",
    "prompt": "Fashion model walking down a city street",
    "size": "1080x1920",
    "seconds": "6"
  }'

Python

# Landscape 16:9 at 1080p
video_landscape = client.videos.create(
    model="veo-3.1-generate-preview",
    prompt="Aerial drone shot of a coastal city at sunset",
    size="1920x1080",
    seconds="8",
)

# Portrait 9:16
video_portrait = client.videos.create(
    model="veo-3.1-fast-generate-preview",
    prompt="Fashion model walking down a city street",
    size="1080x1920",
    seconds="6",
)

JavaScript

// Landscape 16:9 at 1080p
const videoLandscape = await client.videos.create({
  model: "veo-3.1-generate-preview",
  prompt: "Aerial drone shot of a coastal city at sunset",
  size: "1920x1080",
  seconds: "8",
});

// Portrait 9:16
const videoPortrait = await client.videos.create({
  model: "veo-3.1-fast-generate-preview",
  prompt: "Fashion model walking down a city street",
  size: "1080x1920",
  seconds: "6",
});

Prompting Best Practices

To get the best results from Veo 3.1 models:

Be Descriptive and Specific

• Include visual details: colors, lighting, composition
• Specify movement: camera motions, subject actions
• Define style: cinematic, realistic, artistic
• Set mood: atmospheric, energetic, calm

Include Audio Cues

• Dialogue: Use quotes for specific speech (e.g., "This must be the key," he murmured.)
• Sound Effects: Explicitly describe sounds (e.g., tires screeching loudly, engine roaring)
• Ambient Noise: Describe environment soundscape (e.g., A faint, eerie hum resonates in the background)

Specify Camera Movement

• "Camera pans left to reveal..."
• "Slow zoom into subject..."
• "Aerial drone shot descending..."
• "Handheld tracking shot following..."

Good Prompt Example:

A golden retriever puppy playing in a sunlit meadow during golden hour,
camera slowly tracks the puppy running through tall grass,
cinematic depth of field with bokeh effect, warm color grading

Weak Prompt Example:

A dog playing

Model Comparison

Feature	veo-3.1-generate-preview	veo-3.1-fast-generate-preview
Quality	Highest	High
Speed	Standard	Faster
Audio	Rich, natural	High-quality
Cost/Second	$0.40	$0.15
Best For	Production content	Rapid iteration, high-volume
Max Duration	8 seconds	8 seconds
Resolutions	720p, 1080p	720p, 1080p

---

Related Links

• Video API Reference
• Generate Videos Using Veo Guide
• Google Models Documentation
• Pricing Information
• Image Generation Guide
• Best Practices