Hi Google AI team,
We already use Veo 3 via Gemini/Vertex in production and want to plan an upgrade to Veo 3.1 (the version announced with up to 60s videos, clean 1080p, better temporal consistency, and multi-prompt scene transitions).
Could you please clarify the following for Veo 3.1:
-
Availability & timeline
-
ETA for public access in AI Studio, Gemini API, and Vertex AI?
-
Is there an allowlist/preview we can join?
-
-
Model IDs & endpoints
-
Planned model names (e.g.
veo-3-1-generate-*
, fast/standard variants)? -
Any SDK/REST examples we can follow now?
-
-
Capabilities
-
Max duration (60s), resolution (1080p), audio support, multi-prompt & transitions (how to specify prompts/segments)?
-
Improvements to identity/clothing consistency and camera motion controls — any new params?
-
-
Regions, quotas, rate limits
- Supported regions at launch, default quotas, typical latency.
-
Pricing & billing
-
Pricing for Veo 3.1 (per-second or per-request) in Gemini API / Vertex AI.
-
Any differences vs Veo 3 and Veo 3 Fast?
-
-
Migration
-
Backward compatibility: can we reuse our Veo 3 requests with minimal changes?
-
Recommended fallback strategy if Veo 3.1 is temporarily unavailable.
-
Context: We run an API platform that already exposes Veo 3 to customers; having dates, model IDs, and pricing for 3.1 will help us schedule docs, billing, and rollout.
Thanks a lot!