Generated seconds
A route may bill per generated video second. Multiply the charged duration by that route's rate for the selected quality and audio configuration.
Video API cost guide
Eight seconds of output is not a price by itself. First identify the exact provider route, billing unit, resolution, audio setting, and what counts as a billable attempt.
Billing dimensions
A route may bill per generated video second. Multiply the charged duration by that route's rate for the selected quality and audio configuration.
A published count is not automatically one full clip or one second. Use the provider's explicit definition before multiplying.
Use the measured output tokens and video-token price. Text output has its own meter; never substitute the text price for video.
Credits, reserved capacity, discounts, and configuration tiers may add another conversion or eligibility rule. Compare rates only after making the units and platform consistent.
Worked estimate
Suppose a hypothetical video API charges $0.10 USD per generated second for the exact resolution and audio mode you need. This is an illustrative rate, not a quote for any named provider or model.
8 charged seconds × $0.10/second = $0.80
100 × $0.80 = $80.00 before additional charges
120 × $0.80 = $96.00 if those retries are billed in full
Check whether failed attempts or extensions are charged, duration rounding, additional audio, upscaling, input costs, storage, taxes, and regional prices. Count only the charges that the provider actually applies.
Provider-specific examples
Google's Gemini Developer API publishes Veo pricing by generated second, with separate quality, resolution, and audio configurations. The catalog includes veo-3.1-generate-preview and veo-3.1-fast-generate-preview. These exact IDs are listed in Google's preview lifecycle documentation. They are not interchangeable with similarly named Google Cloud/Vertex AI routes.
Sources: Gemini API pricing and Gemini model lifecycle.
Google Cloud's Veo pricing table uses USD per count for listed configurations. This guide does not assume one count is one second or one complete video: verify the exact route and SKU's billing definition before turning that row into a clip estimate.
The exact gemini-omni-1.2-flash route has separate published Standard PayGo meters: $1.50 USD / 1M input tokens, $9.00 USD / 1M text output tokens, and $12.00 USD / 1M video output tokens. A clip-cost quote needs the video tokens actually billed for that exact generation configuration; do not convert the $12 token rate to a universal second rate.
Source: Google Cloud pricing — Gemini Omni.
Availability matters
OpenAI's Videos API and Sora 2 API models were shut down on September 24, 2026. Historical price tables may still circulate, but these routes are not current API purchasing options. Check the official deprecation record before using an older comparison.
For a new project, confirm the provider, exact model ID, lifecycle, region, access requirements, and official pricing date. The regular token calculator is not a universal video-cost estimator; use it only for billing dimensions it supports.
Official Google Gemini, Google Cloud, and OpenAI sources above were reviewed November 7, 2026. Model detail pages expose the catalog's available price fields and provenance. Pricing and access change; recheck the linked provider page before committing a production budget.