← Video
Initial research 2026-09-21Google DeepMind

Veo 3.1

Provider status: preview · Identity verified 2026-10-15 · Official catalog

Inspect the documented capabilities, access conditions and sources for your task.

Initial research: 2026-09-21. Check current terms before choosing.

Automated fact checks · 1 fields

AI extraction with a second evidence validation pass. These dates apply to the listed fields.

  • identity · 2026-10-15 · Source

Updated

The facts

Version
3.1
Price
$0.4 per second of video. Usage rate. Settings and included usage can change the charge.
API
✓Supported
Open weights
✕Not supported
Commercial use
Google's terms allow commercial use of video generated on paid subscriptions. Caveat noted by reviewers: AI-generated content is generally not copyrightable in the US, so competitors could legally reuse your output.
Max Clip Length
~Partly

Native generations are 4, 6 or 8 seconds; scene extension chains these to up to 60 seconds of continuous footage, described as the longest of any major model via extension.

Source for Max Clip Length (opens in a new tab)
Audio Generation
✓Supported

Native synchronised audio in the same call — dialogue, ambience and effects locked to on-screen action, at 48 kHz speech. Reviewers call this Veo's remaining moat.

Source for Audio Generation (opens in a new tab)
Image To Video
✓Supported

Generates 1080p from text or image prompts.

Source for Image To Video (opens in a new tab)
Character Consistency
~Partly

Reviewers rank Kling 3.0 and Seedance ahead for multi-shot subject consistency; Veo's strength is audio sync.

Source for Character Consistency (opens in a new tab)

Worth knowing

deprecation

Google shut down veo-3.0-generate-001, veo-3.0-fast-generate-001 and veo-2.0-generate-001 on 30 June 2026.

2026-06-30Google AI for Developers

Commercial risk

No litigation found in fetched sources. Google is still shipping incremental 3.x updates with no announced Veo 4 roadmap.

From the publisher

Automatic source checks. These do not change the verification date of the facts above.

Sources

Beyond feature support

Quality evidence.

How well a model performs depends on the task, version, and settings.

We have not published a reviewed quality evaluation for these versions yet. Feature support and price do not establish output quality.

Try the same small task with each option. Check accuracy, consistency, and how much editing you need to get a usable result.

Start a conversation.