Veo 3.1 — Google DeepMind Video AI with Native Audio
Google DeepMind's Veo 3.1 on SeedDance. Create cinematic clips with optional native audio, stronger prompt adherence, and improved image-to-video — from text or image inputs, in 4, 6, or 8 seconds, up to 1080p.
Available on SeedDance platform
Cinematic Video from Text or Image
Released in October 2025, Veo 3.1 evolves Veo 3 with stronger prompt adherence, richer native audio, and improved image-to-video quality. On SeedDance you can use Veo 3.1 for text-to-video and image-to-video with optional audio, up to three reference images, durations of 4 / 6 / 8 seconds, and 720p or 1080p output.
Native Audio Generation
Veo 3.1 can generate video and audio together — dialogue, ambient effects, and music synthesized with the visuals for tighter sync, without a separate audio pipeline.
Stronger Prompt Adherence
Compared with Veo 3, Veo 3.1 follows complex instructions more reliably — camera angles, lighting, pacing, character behavior, and scene composition.
Improved Image-to-Video Quality
Animate still images with more natural motion and better continuity. On SeedDance you can provide up to three reference images to guide the generation.
Real-World Physics & Realism
Stronger realism in fluids, lighting, object interactions, and human motion — suited to cinematic drafts and product-style clips.
What You Can Do with Veo 3.1 on SeedDance
Core Veo 3.1 capabilities available in the SeedDance generator — with clear specs for duration, resolution, and audio.
SeedDance Feature Set for Veo 3.1
Capabilities exposed in the SeedDance AI video generator for Veo 3.1.
Text-to-Video Generation
Transform detailed text prompts into cinematic clips. Veo 3.1 understands language, spatial relationships, and temporal flow for coherent results.
Image-to-Video Generation
Animate still images with natural motion and optional audio. Upload up to three reference images and describe the desired motion and camera work.
Optional Native Audio
Generate dialogue, ambience, and effects with the video when audio is enabled — timed to on-screen action without a separate dubbing step.
Prompt-Driven Camera Control
Describe pans, tilts, zooms, tracking shots, and cinematic transitions in your prompt — Veo 3.1 follows cinematography cues with strong adherence.
Physics-Aware Motion
More believable fluids, object interactions, lighting, and human motion for clips that feel grounded rather than floaty or inconsistent.
4 / 6 / 8 Second Duration
Pick 4, 6, or 8 seconds per generation for social clips, product demos, and short story beats.
720p / 1080p Quality
On SeedDance, Veo 3.1 outputs at 720p or 1080p with 16:9 or 9:16 aspect ratios (auto available for image-to-video).
SynthID Content Authentication
Veo outputs are marked with Google's SynthID watermarking technology, supporting transparent identification of AI-generated media.
Frequently Asked Questions
Everything you need to know about Veo 3.1 and how to use it on SeedDance.
Start Creating with Veo 3.1 Today
Try Veo 3.1 on SeedDance — text-to-video and image-to-video with optional native audio, 4 / 6 / 8 second clips, and up to 1080p.