Google Veo
by Google (Google DeepMind)
Google’s video generation family with native audio. Closed models, available only through Google products and APIs (no open weights).
Status (Gemini API docs, 2026-09-30)
- Veo 3.1 (
veo-3.1-generate-preview) and Veo 3.1 Lite (veo-3.1-lite-generate-preview): both listed as Preview. Veo 3.1 supports native audio, video extension and frame-specific (first/last frame) generation. Launched October 2025 (date from earlier knowledge; not re-fetched). - Gemini Omni Flash (
gemini-omni-1.1-flash, Stable): fast video generation, editing, keyframe interpolation and extension with native audio. The video docs now recommend it as the default, keeping Veo 3.1 for scene extension, last-frame control and legacy pipelines. - Retired:
veo-3.0-generate-001,veo-3.0-fast-generate-001andveo-2.0-generate-001were scheduled for shutdown on 2026-06-30 (replacementveo-3.1-*-preview).
Pricing
Usage-based, per second of generated video on the Gemini API (Lite is cheaper than the standard tier; Omni is token-priced). Current prices: see https://ai.google.dev/gemini-api/docs/pricing
Access
Gemini API and AI Studio, Vertex AI, Google Flow and the Gemini app; third-party hosts such as kie-dot-ai and fal-dot-ai resell API access (their prices are not recorded here).
Features and limits
Text-to-video, image-to-video, extension, first/last frame. Duration and resolution limits were not stated in the docs page fetched: unverified. Outputs are watermarked with SynthID (Google’s standard; not re-checked for video this session).
Related
ai-video-generation-tools, openai-sora, kling, seedance, nano-banana
Sources
- https://ai.google.dev/gemini-api/docs/models (accessed 2026-09-30)
- https://ai.google.dev/gemini-api/docs/video (accessed 2026-09-30)
- https://ai.google.dev/gemini-api/docs/deprecations (accessed 2026-09-30)
- https://ai.google.dev/gemini-api/docs/pricing (accessed 2026-09-30)
Merged from google-veo-3-dot-1 (2026-10-01, unverified)
Unique content from an unreviewed Dec 2025 draft; all items below are unverified and may be outdated.
- Output: 720p/1080p, 16:9 and 9:16, clips of 4, 6 or 8 seconds, extendable via “Extend”.
- Creative controls: Ingredients to Video (multiple reference images), Frames to Video (start/end image bridging), Add/Remove Objects.
- Access also via Google Vids; Google Vids avatars claimed 5x more realistic.
- Draft pricing claims in the unreviewed draft were removed (prices go out of date).
- Use cases listed: ads from product shots, vertical social clips, film prototyping.