Veo 3.1 generates video with native audio through the Gemini API, priced per second ($0.40 for 720p/1080p Standard, up to $0.60 for 4K; a cheaper Fast tier from $0.10). Google's docs now recommend Gemini Omni Flash as the default for most video generation, positioning Veo 3.1 for specific needs like scene extension or last-frame control. Check which model your workflow actually needs before assuming Veo is the default choice.
Other good fits
Projects that want native audio generated alongside the video in one pass.
+ Pros
Read pros
Veo 3.1 generates video with native audio through the Gemini API, and supports specific production-oriented features: extending an existing generated scene, controlling the last frame precisely, and image-based direction through the generateContent API. As of the documentation checked, Google now recommends Gemini Omni Flash as the default choice for most video generation tasks, positioning Veo 3.1 specifically for cases needing those specialized capabilities or integration with a legacy Veo-based pipeline.
Pick it when…
You need scene extension, last-frame control, or another Veo-specific capability that Omni Flash doesn't offer.
You're integrating with an existing pipeline already built around Veo.
Native audio generated alongside the video in one pass matters for your workflow.
− Cons
Read cons
Google's own current guidance points most users toward Gemini Omni Flash rather than Veo 3.1, so defaulting to Veo without checking current recommendations risks using a more specialized (and not necessarily cheaper) tool than you need. Availability and limits vary by product surface and are stated to be subject to change, so don't assume today's limits hold indefinitely; recheck before scaling up a production integration.
Pick something else when…
You just need general-purpose video generation. Check Gemini Omni Flash first, per Google's own current recommendation.
You want a different cloud video vendor to compare.Kling or Runway are worth checking against Veo's pricing and features.
You want a self-hosted, open-weight alternative instead of a hosted API.Wan runs on your own hardware.
Avoid if
You need the lowest possible cost; check Omni Flash's token-based pricing against Veo's per-second rate for your specific output length and resolution.
Context
Check current Google documentation before assuming Veo is the right default: as of this review, Gemini Omni Flash is recommended for most video generation, with Veo 3.1 reserved for specific needs. Model recommendations and pricing structures both change; verify which model fits your actual requirement before building around either one.
Requirements
This is a hosted API call through the Gemini API; you need a Google AI API key and network access, not local hardware. Availability and output limits vary by product surface (API vs. consumer app) and can change, so confirm current limits for your specific access point.
~$0.10/second (token-based: 5,792 tokens/second of 720p at $17.50/1M tokens)
Veo bills directly per second; Omni Flash applies its standard token pricing to video output, which works out to a comparable rate at 720p but scales differently at other resolutions or durations. Compare both for your specific output length and resolution rather than assuming one is always cheaper.
First task
Documentation-based suggestion — not run here.
Read first task
Compare Veo 3.1 and Omni Flash on the same short clip.
Get Gemini API access and pick a short, specific scene to generate (a few seconds, no real people's likeness).
Generate it once with Veo 3.1 Standard and once with Gemini Omni Flash, using the same prompt.
Compare quality, whether audio is included and useful, and the actual cost of each based on duration and resolution.
Decide which fits your project's actual need, rather than defaulting to whichever name is more familiar.
Done when you've generated the same prompt through both models and can point to a specific reason (a capability or cost difference) for choosing one over the other.
Links & evidence
Documentation-based · not hands-on · Sep 23, 2026
Rests on Google's Gemini API video generation guide and pricing documentation, checked 23 September 2026. No API key was used and no video was generated for this review.
Gemini API video generation · Sep 23, 2026 — Veo 3.1 capabilities (scene extension, last-frame control, native audio), Gemini Omni Flash as recommended default.
Gemini API pricing · Sep 23, 2026 — Veo 3.1 Standard/Fast/Lite per-second rates; Gemini Omni Flash token-based video pricing.
Generate video scenes from text prompts using Google's cloud model.
Include generated audio when a scene needs sound or dialogue.
Define shots, rights, captions, and review steps before publishing results.