Veo 3.1 pricing comparison across Standard, Fast, and Lite tiers. See exact per-second costs for 720p, 1080p, and 4K video generation with the new Google DeepMind model.
Veo 3.1 introduces three pricing tiers — Standard, Fast, and Lite — replacing Veo 2.0's single flat rate. The new structure lets you trade speed and resolution for cost, making the model accessible across more budgets.
All three tiers include native audio generation, meaning dialogue, ambience, SFX, and BGM are baked into the price — no separate audio processing costs.
The complete cost breakdown across the three tiers and three resolutions:
Veo 2.0 charged a flat $0.35/second regardless of resolution — simple but inflexible. Veo 3.1's tiered system means you can now generate at Lite pricing for rapid prototyping ($0.05/s) and only scale up to Standard for final renders.
For a typical workflow of 10 iterations at 8 seconds each on Fast 1080p, the cost is about $9.60 — roughly 30% cheaper than the same workflow on Veo 2.0. If you use Lite for drafts and Standard only for finals, the savings grow further.
Standard is best for final client delivery, broadcast content, and any work where maximum visual fidelity matters. The 4K option at $0.60/s makes feature-quality output accessible without a studio budget.
Fast hits the sweet spot for most creators: clean 1080p at $0.12/s lets you iterate through shot variations without watching the meter. Use Fast for daily content and social video pipelines.
Lite is the prototyping workhorse. At $0.05/s for 720p, you can rapid-fire test prompts, validate camera choreography, and block out scenes before committing to a higher tier. Lite 1080p at $0.08/s is also strong enough for short social clips where perfection isn't required.
A 30-second social ad in Fast 1080p costs $3.60. A 60-second narrative scene in Standard 1080p costs $24.00. A one-minute cinematic short with dialogue and layered audio in Standard 4K costs $36.00.
Compared to traditional production costs, this is orders of magnitude cheaper. The tiered model also means you do not pay a premium for fast iterations — use Lite for drafts, Fast for review cuts, and Standard for the final export.