Google Vids API

October 7, 2026 (November 16, 2026)

Independent third-party service, see Legal

This is a third-party API for Google Vids by Google. It uses your Google AI subscription: Free, Plus $4.99 (or $9.99 with 2 TB of storage), Pro $19.99, Ultra $99.99 or Ultra $199.99 a month (as of November 2026). Through a REST API and a hosted MCP server, it makes Gemini Omni 1.2 Flash videos of up to 15 seconds with sound at 720p or 1080p, extends, upscales and edits them, and makes images, speech and talking avatars with 30 voices. Google publishes no Google Vids API, and Vids asks for no captcha and has its own monthly Omni allowance, separate from Google Flow credits.

Models

Model What it makes Notes
Gemini Omni 1.2 Flash Video with sound and speech from text, a start image, or up to 3 references that mix your images and avatars 3 to 15 seconds, 720p or 1080p, landscape or portrait, extend by up to 10 seconds at a time, on clips of up to 24 seconds, to about 34 seconds, upscale to 1080p, edit up to 15 seconds by prompt
Avatars Avatars: a picture and one of 30 voices From a description, your own photo, a Vids image or another avatar, changed in words or with a logo, up to 3 in one Omni video for a dialogue, kept until you delete them
Talking avatar video A talking head that reads your script in the avatar’s voice Up to 30 seconds, 720p landscape, with sound
Gemini voices Speech from a script WAV in 30 voices, scripts up to 2,500 characters, audio tags like [whisper]
Vids images Images from text About 1 megapixel, 16:9, 1:1 or 9:16, in 7 styles, for references and avatar pictures
More Uploads of JPEG, PNG, WebP or MP4 up to 50 MB, downloads, and avatars made in the Vids web app  

Google added Omni to Vids in September 2026 and moved it to Omni 1.2 Flash, with clips of up to 15 seconds, on November 5, 2026. The Vids web app calls it just “Omni”. Make a single clip at 1080p directly, for the same seconds as 720p. For a clip you will extend or edit, work at 720p and upscale once at the end: a 1080p clip is an upscaled 720p one, and each extend at 1080p processes the whole clip again and over-sharpens faces (see the comparison). Vids images are basic, with no reference images and no larger size. For better images use the Google Flow API: Nano Banana with reference images, at up to 4K.

Every video carries Google’s invisible SynthID watermark and a visible Gemini ✦ in a corner, and the API returns Google’s file unchanged. The free watermark remover in our GitHub examples takes the visible ✦ off clips as generated and their upscales (details).

MCP server

The hosted Google Vids MCP server gives Claude Code, Cursor, VS Code and other AI assistants 18 tools for videos, extends, upscales, edits, images, uploads, avatars, talking avatar videos and speech, with your useapi.net API token and nothing to install. It is listed in the official MCP Registry as net.useapi/google-vids, and the google-vids-api repo also adds it in one step as a Claude Code plugin or a Gemini CLI extension.

Pricing

What one video costs on Google’s Gemini API and on each Google AI subscription (as of November 2026). Vids has no credits: each subscription comes with seconds of video, a number of images and a number of AI avatar videos a month, listed on Google’s Gemini in Google Vids page, so a video costs its share of the subscription. On a subscription a 1080p video uses the same seconds as a 720p one, while Google’s API charges 50% more for it.

Model Official API Ultra $199.99/m
10,000 s/mo
Ultra $99.99/m
2,500 s/mo
Pro $19.99/m
500 s/mo
Plus $4.99/m
60 s/mo
Omni 1.2 Flash
720p, 15 s
$1.04
$0.07/s
$0.30 / 15 s
666 videos/mo
$0.60 / 15 s
166 videos/mo
$0.60 / 15 s
33 videos/mo
$1.25 / 15 s
4 videos/mo
Omni 1.2 Flash
1080p, 15 s
$1.56
$0.104/s
$0.30 / 15 s
666 videos/mo
$0.60 / 15 s
166 videos/mo
$0.60 / 15 s
33 videos/mo
$1.25 / 15 s
4 videos/mo
Omni 1.2 Flash
720p, 10 s
$0.70
$0.07/s
$0.20 / 10 s
1,000 videos/mo
$0.40 / 10 s
250 videos/mo
$0.40 / 10 s
50 videos/mo
$0.83 / 10 s
6 videos/mo
Vids image no API $0 / 0 s
1,000 images/mo
$0 / 0 s
300 images/mo
$0 / 0 s
30 images/mo
$0 / 0 s
≤50 images/mo
Avatar no API $0 / 0 s
1 image
$0 / 0 s
1 image
$0 / 0 s
1 image
$0 / 0 s
1 image
Talking avatar
video, up to 30 s
no API $0 / 0 s
1 avatar video
$0 / 0 s
1 avatar video
$0 / 0 s
1 avatar video
$0 / 0 s
1 avatar video
Speech no API $0 / 0 s
unlimited
$0 / 0 s
unlimited
$0 / 0 s
unlimited
$0 / 0 s
1 of 50 credits

Scroll the table sideways for more plans →

  • A free Google account makes avatars at no cost, and in our tests (November 2026) it had 10 AI avatar videos and 50 Gemini uses (speech) a month, but no video seconds or images in Vids, so Omni videos and images skip it.
  • A video uses its length in seconds, at the same rate for 720p and 1080p, and references or avatars add nothing. An extend uses only the seconds it adds, an upscale uses the clip’s seconds again, and an edit uses at most the seconds it edits (in our tests a 15-second edit used 10). A request Google refuses uses nothing.
  • Video seconds are shared by everyone on a Google family plan, while images are counted per account. Allowances reset on the 1st of the month at 00:00 Pacific time.
  • A talking avatar video uses 1 of the account’s AI avatar videos and 1 Gemini use for its speech, and no video seconds. Google counts it even when its safety check refuses the finished video. Speech uses 1 Gemini use.
  • Google lists AI voice-over as unlimited on Google AI Pro and Ultra, and as 1 credit per voice-over on a free account and on Plus, out of 50 credits a month shared with other Vids actions (Google’s limits). The API follows the speech counter Google reports for each account: in our tests (November 2026) an account in an Ultra $199.99 family plan, not the plan owner, had 50 a month. GET /accounts/email shows it.
  • GET /accounts/email shows what each account has left. When an allowance runs out, the job fails with 429 and the reset time, and the account takes no more jobs of that kind until then.
  • The seconds and images per plan are Google’s figures, and our tests on Ultra $199.99 match them. Plan prices are Google’s US prices.

Vids seconds are separate from Google Flow API credits, so one Google account connected to both makes Omni videos in each, on top of each other.

One Google account, four APIs

The same Google AI plan powers several useapi.net APIs, each from its own allowance, so one Google account connected to all of them gets the most out of the plan:

  • Google Flow API: Veo 3.1, Omni 1.2 Flash and Omni 1.1 Flash video, Nano Banana images up to 4K, from Flow credits.
  • Google Vids API: Omni 1.2 Flash video up to 15 seconds, 30-second talking avatar videos from a photo, avatars in 30 voices and text to speech, from the Vids allowance, which does not use Flow credits.
  • Gemini Notebook API (NotebookLM API): chat with sources, Deep Research and every Studio output, from the Gemini Notebook usage limits.
  • Flow Music API: full songs with Lyria in Google Flow Music, on the Flow Music plan your Google AI subscription includes (Starter on Plus, Plus on Pro, Member on Ultra) or on Flow Music’s own plans.

Each API is set up on its own setup page, and one useapi.net subscription covers all of them.

Start using the API

A single $15/month subscription covers every API on useapi.net — one token, all services. You connect up to 3 accounts per service, each with its own website subscription, and useapi.net handles the REST API, load balancing, and polling. Need more accounts? Each extra subscription is $10/month and adds 3 more accounts per service.

Subscribe — $15/month Pay with crypto

Pricing and volume options — full refund within the first 14 days if you have fewer than 50 successful generations. Already subscribed? Change the quantity in the Stripe Customer Portal instead of buying again.

Run several Google Vids accounts

Each useapi.net subscription connects 3 Google accounts, up to 100 in total. Video and image requests don’t need email: each job goes to an account that is healthy, has a free slot within its maxJobs (1 to 10, default 3) and has allowance left for that kind of job. A job Google refuses because the account’s allowance is used up moves to another account with allowance left and runs there. When every account is busy you get 429, so wait and send it again. When every account has used up its allowance, the 429 comes at once with retryAt, the earliest reset time. Pass email to pin a job to one account. A job that uses an upload, an avatar or an earlier clip runs on the account that holds it. Accounts of one family plan share their video seconds, so adding them runs more jobs at once and adds images, but no seconds.

Every job can post its finished record to your webhook with replyUrl, or you poll GET /jobs/jobid, so a video, which takes from about 20 seconds to over 2 minutes for an extend or an edit, needs no open connection. GET /jobs shows what each account is running. Download results soon: videos stay available for 24 hours and images for a few hours (how long files last).

How useapi.net works has tips for keeping your accounts healthy.

Questions

Does Google Vids have an API?

Through useapi.net, yes. Google has published no Google Vids API (as of October 2026), and the useapi.net Google Vids API makes Gemini Omni 1.2 Flash videos of up to 15 seconds, images and talking avatars in Vids over REST, on your Google AI subscription. The MCP server gives the same tools to Claude Code, Cursor and other AI assistants, with nothing to install.

How much does the Google Vids API cost?

Through useapi.net, a 10-second Omni 1.2 Flash video uses about $0.20 of Google AI Ultra $199.99, against about $0.70 at 720p on Google’s Gemini API (as of November 2026). On the same plan a 15-second 1080p video uses about $0.30, against $1.56 on the Gemini API. Images and avatars use no video seconds. The price table shows every plan.

Is there a Google Vids API key?

Through useapi.net, one API token is the only key you need. Connect your Google account once with the automated setup: you sign in with Google in a remote browser, with no cookies to copy, no Google Cloud project, no Gemini API key and no captcha provider. If Google ever signs the account out, you get an email with a link to re-add it.

Can the Google Vids API make talking AI avatars?

Through useapi.net, yes, while Google offers Vids avatars only in its web app: POST /avatars pairs a person’s picture, drawn or from your own photo, with one of 30 Gemini voices. The avatar reads a script of up to 30 seconds as a talking-head video, or up to 3 avatars speak the lines you write in a video prompt. The AI influencer tutorial builds a clip step by step.

Setup Google Vids
Google Account Setup Scripts for a clean, single-use Google sign-in (Windows, macOS, Linux)
LLM-friendly API spec Feed this to your LLM to build integrations
Claude Code skill A drop-in useapi.net skill for Claude Code, and a rule for Cursor
Postman collection (November 6, 2026)
MCP Server Make Vids clips, images, talking avatars and speech from Claude Code, Cursor, VS Code and other MCP clients
Agent skill: google-vids-video Claude Code and Codex make Omni 1.2 Flash clips and talking avatars for you: npx skills add useapi/google-vids-api --skill google-vids-video
GitHub examples Node.js and Python: voiced-avatar influencer clips, an n8n workflow, an agent skill for Claude Code and Codex, and a remover for the visible Gemini ✦
n8n template: Talking Avatar App Describe or pick a presenter, choose a voice, write the lines and get a talking avatar video of up to about 34 s, ready to import

Articles:

Examples:

Videos on YouTube:

Developer Community:


Table of contents