Google Flow Credits vs Veo API Pricing: What Buyers Should Separate

Short answer: Google Flow credits are a buyer-facing subscription and workspace access meter for using Flow. Gemini API and Vertex AI Veo billing are developer and Cloud routes that charge against an API key, project, or Cloud account. Do not assume ordinary Flow credits include Gemini API or Vertex AI Veo usage unless a Google product page or contract explicitly says that route shares the same credits.

What Flow credits buy

Google Flow is the creative-studio route. A buyer subscribes to a Google AI plan, receives a monthly or daily Flow credit allowance, and spends those credits inside Flow for generations, edits, upscaling, and related creative actions. The unit is not a dollar-per-second API meter. It is an in-product allowance tied to the Flow experience.

Google's subscription page places Flow credits inside consumer Google AI plans. Google AI Plus is listed at $7.99 per month with 200 Google Flow credits, Google AI Pro at $19.99 per month with 1,000 Flow credits, and Google AI Ultra starting at $99.99 per month with 10,000 or 25,000 Flow credits depending on the Ultra tier. Flow credit help also says people without a paid Google AI plan can receive daily free Flow credits for trial use.

The spending rule is per generation, not per prompt. Google warns that one request can create multiple generations. Flow credit costs also vary by model and mode: Veo 3.1 Lite, Fast, and Quality can have different credit costs, Ultra subscribers can receive lower costs on some Fast or Lite generations, and 4K upscaling is treated differently from basic 1080p upscaling.

That is why Flow credits work best as a creator planning budget. They answer questions like how many Flow drafts, revisions, upscales, and short clips a team can attempt inside Google's creative interface before it has to wait for refresh, upgrade, or buy eligible additional AI credits.

Where top-ups and refresh rules fit

Flow plan credits and purchased AI credits are not the same operational promise. Flow plan credits refresh on a billing cycle, unused monthly credits do not roll over, and free daily credits do not roll over. Purchased AI credits can have their own expiration terms, family-plan controls, and checkout eligibility.

Google's AI credit purchase help says one-time AI credit purchases are available through Google One and select Google apps for Google AI Pro and Google AI Ultra plans, are not available for Google AI Plus plans, and are not available in Japan. Flow credit help also points subscribers in supported regions to additional AI credit purchases, so the practical rule is to treat checkout eligibility as plan-, country-, and account-specific.

For individual creators, top-ups are a pressure valve when a Flow project runs past the included allowance. They are still Flow and Google AI account credits, not a replacement for API billing. Buying more AI credits should not be read as funding Gemini API calls or Vertex AI jobs unless the specific developer product says so.

How Workspace changes the route

Workspace adds an administrator-managed version of the Flow credit story. Google Workspace admin help says qualifying Workspace editions get Google Flow at no extra charge with 100 monthly AI credits, and AI Ultra Access adds more AI credits, higher Flow limits, Veo access, and premium features such as Ingredients to Video.

For AI Ultra Access, Google describes 25,000 monthly credits for each licensed user to use with Flow, with credits unique to Flow and not affected by other Google Workspace with Gemini usage. Those credits refresh on the first day of each month and do not roll over.

Workspace overages are also an admin decision, not a self-serve creator assumption. Google says credit overages for AI Ultra Access users are disabled by default, can be capped per user, and are billed only for additional credits used. If overages are disabled or a user reaches the cap, Flow and Whisk actions stop until the cap increases or credits refresh.

This makes Workspace the right route when the buyer needs centralized procurement, admin controls, auditability, and organization-level rules. It is not the same as using a developer API, even when the underlying model family is Veo.

How Gemini API and Vertex AI billing separate

Gemini API is the developer route for programmatic Veo use. Google's Gemini API docs describe Veo 3.1 as accessible programmatically, with asynchronous video-generation operations that you poll until complete. The Gemini API pricing page lists Veo models on the paid tier and prices them per second in USD, with no free tier for Veo.

On the Gemini API pricing page, Veo 3.1 Standard video with audio is priced at $0.40 per second for 720p and 1080p and $0.60 per second for 4K. Veo 3.1 Fast is listed at $0.10 per second for 720p, $0.12 for 1080p, and $0.30 for 4K. Veo 3.1 Lite is listed at $0.05 per second for 720p and $0.08 for 1080p, with 4K not supported. Google also notes that you are charged only if the video is successfully generated when an audio-processing issue prevents generation.

Vertex AI is the Google Cloud route. It is usually the better fit when the buyer needs Cloud projects, IAM, enterprise controls, regional endpoints, and Cloud billing. Google's Cloud pricing separates several Veo rows by model, output, and resolution, including video-with-audio and video-only rates. That separation matters because Cloud usage can be billed differently from the consumer Gemini API page and from Flow credits.

The core boundary is simple: Flow is a subscription access product with credits inside an app. Gemini API and Vertex AI are build routes with metered developer or Cloud billing. The fact that all three can expose Veo does not make the credit balances interchangeable.

Region and availability checks

Flow availability is country-, age-, language-, and feature-dependent. Google's Flow availability page says Flow is available to users ages 18 and older in supported countries and territories, recommends English prompts for best results, and notes that product feature availability can differ by country. It also lists uploaded and generated video edit restrictions for the EEA, India, Switzerland, the United Kingdom, and some US states.

Workspace adds another gate: admins control whether Google Labs tools are on for users, and education environments can have different defaults. AI Ultra Access also has its own purchase availability and terms because Flow is an additional service outside some core Workspace services.

Vertex AI has a different location question. Google Cloud says Vertex AI has no global location and that requests must use supported regional endpoints. For Veo on Vertex AI, a location is the region you specify in the request to control where data is stored at rest, so a team should check both model support and region requirements before estimating cost.

Which route should a buyer choose

Use Flow when the job is interactive filmmaking: prompting, trying scenes, using a creative workspace, organizing assets, and making short videos without building an app. In that route, budget around plan price, included Flow credits, credit costs per generation, top-up eligibility, rollover rules, and regional feature limits.

Use Gemini API when the job is developer integration and the team wants to call Veo from code without adopting the broader Vertex AI governance layer. In that route, budget by model, output seconds, resolution, audio behavior, rate limits, success-only charging rules, and API key controls.

Use Vertex AI when the job belongs in Google Cloud governance: project-level billing, IAM, regional endpoint control, enterprise procurement, monitoring, and integration with other Cloud workloads. In that route, treat Veo as Cloud consumption, not as a Flow subscription benefit.

The final pricing check is not just the visible monthly plan price. Confirm the route first, then verify the unit: Flow credits per generation, Google AI or Workspace top-ups and overages, Gemini API per-second pricing, or Vertex AI Cloud pricing. Mixing those units is the fastest way to underestimate a Veo budget.