Google Veo 3.1

Cinematic video with native audio and reference-image control

About model

On Together AI, Google Veo 3.1 costs $0.08 per video.

Veo 3.1 is Google's video generation model, building on Veo 3 with stronger prompt adherence and improved audiovisual quality, particularly when turning images into video. It generates video with synchronized native audio, including natural dialogue, ambient sound, and sound effects. Up to three reference images of a character, object, or scene guide generation to keep subjects and style consistent across shots, and a supplied first and last frame lets the model generate a seamless transition between them, complete with audio. Output is available at 720p or 1080p in landscape and vertical formats. Available on Together AI.

‍

Synchronized Audio

Native

Dialogue, ambient sound, and effects generated with the video

Reference Images

Keep characters, objects, and style consistent across shots

Output Resolution

1080p

Landscape and vertical formats

Model key capabilities

Native Audio: Generates natural dialogue, ambient sound, and sound effects synchronized with the visuals
Reference-Guided Consistency: Up to three reference images keep characters, objects, and style consistent across multiple shots
Frame-to-Frame Transitions: Generates a seamless transition between a supplied first and last frame, with accompanying audio
Available on Together AI: Serverless access through the Together AI API

API usage

cURL
Python
Typescript

Endpoint:

google/veo-3.1

curl --request POST \
  --url https://api.together.ai/v2/videos \
  --header "Authorization: Bearer $TOGETHER_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "google/veo-3.1",
    "prompt": "some penguins building a snowman"
  }'

from together import Together

client = Together()

# Create a video generation job
job = client.videos.create(
    prompt="A serene sunset over the ocean with gentle waves",
    model="google/veo-3.1"
)

import Together from "together-ai";

const together = new Together();

async function main() {
  // Create a video generation job
  const job = await together.videos.create({
    prompt: "A serene sunset over the ocean with gentle waves",
    model: "google/veo-3.1"
  });

Model card
Architecture Overview:
• Google video generation model building on Veo 3
• Text and image input with video output, including synchronized native audio
• 720p and 1080p output in 16:9 and 9:16 formats

Training Methodology:
• Developed by Google as the successor to Veo 3, with improvements to prompt adherence and audiovisual quality in image-to-video generation

Performance Characteristics:
• Google reports more lifelike textures and lighting, smoother motion, and fewer visual artifacts than Veo 3
• Audio generation follows the scene, covering dialogue, ambience, and sound effects
‍
Prompting
Together AI API Access:
• Access Google Veo 3.1 via Together AI APIs using the endpoint google/veo-3.1
• Authenticate using your Together AI API key in request headers
• Create a video request from a text prompt, optionally adding reference images or a first and last frame, then poll until the video completes
• Available on Together AI with serverless access
‍
Applications & use cases
Advertising & Brand Content:
• Produce product and brand clips with consistent subjects anchored by reference images
• Generate vertical and landscape variants for social and web placements
• Add dialogue and sound design in the same generation

Film & Storyboarding:
• Previsualize scenes with matching ambience and dialogue
• Bridge two key frames with a generated transition
• Keep characters on model across a sequence of shots

Social & Creator Content:
• Animate still images into short clips with audio
• Generate explainer and social video from text descriptions
• Iterate quickly on creative directions from one reference set
‍

Model specifications

Model data

Model provider
Google
Type
Video
Main use cases
Text-to-Video
Image-to-Video
Resolution/Duration
720p-1080p
Deployment
Serverless
Endpoint
google/veo-3.1
Price
$0.08 / video
Input modalities
Text
Image
Output modalities
Video

Released
October 14, 2025
Category
Video

Run in Playground

Quickstart docs

Deploy Model

Google Veo 3.1

About model

API usage

Model card

Prompting

Applications & use cases