Veo 3.1

A core strength is native audio generation. Veo 3.1 produces ambient sound, sound effects, and dialogue synchronized with the video in a single model pass. Many tools still output silent clips that need a separate audio step. Kling O3 and Wan 3.0 also generate audio, so test them on the same brief; Veo 3.1 is a dependable pick when audio-visual coherence and motion quality matter most.

In Masonry, a single generation produces 4, 6, or 8 seconds of video. Google's API also offers clip extension, but the Masonry route does not expose it today, so build longer pieces from several shots that share reference images.

In Masonry, Veo 3.1 generates at 1280×720 or 1920×1080 in 16:9 (landscape), and 720×1280 or 1080×1920 in 9:16 (portrait). There is no 1:1 or 4:3 output on this route.

Yes. Veo 3.1 supports image-to-video, accepting a reference image at 720p or higher resolution as the opening frame. The model animates from that starting frame, which is useful for bringing existing brand photography, product shots, or concept art to life without re-prompting the entire visual from scratch.

Veo 3.1 generates dialogue with tight lip-sync, so spoken audio tracks the character's mouth movements closely rather than looking dubbed. Google notes that perfectly consistent spoken audio is still an area of active development, but generating speech in the same pass as the video is a meaningful improvement over layering audio in afterward for ad creative and social content with talking characters or voiceovers.

Veo 3.1 generates its audio (ambient sound, sound effects, and dialogue) in the same pass as the video, so it arrives already synced to the visuals rather than layered on afterward. You can describe the desired ambience, effects, and dialogue in the prompt and the model will synthesize audio that fits the described scene.

Commercial use of Veo 3.1 outputs is permitted for users on Vertex AI or Gemini Enterprise subscriptions under Google's API terms. This covers advertising, marketing materials, branded content, and client deliverables. All outputs include SynthID watermarking for provenance tracking, which does not restrict commercial use.

No visible watermark is added. Every output carries an invisible SynthID digital watermark embedded in the video data. SynthID survives common editing operations including color grading, cropping, and compression, and it functions as a machine-readable provenance record rather than a visible brand mark.

Veo 3.1 is billed per second of generated video on the Vertex AI and Gemini APIs, with a lower-cost Veo 3.1 Fast tier for teams that prioritize speed over maximum quality. Per-second API rates change over time, so check Google's current Vertex AI pricing for exact figures. For non-API access, Google AI Pro and Ultra subscriptions bundle generation credits through Google's consumer apps.

Veo 3.1 is notably strong at fast-motion content (athletes, vehicles, pouring liquids, and sizzling food), which is why it works well for sports, beverage, and food advertising. The model maintains subject coherence and realistic physics through rapid motion, where competing models sometimes produce flickering or distorted subjects.

Describe three things explicitly, the visual scene, the motion happening in the frame, and the audio you want (ambient, effects, or dialogue). Veo 3.1 generates all three natively, so including audio intent in the prompt gives you a more complete clip in one pass. Camera movement cues (dolly, pan, handheld) also carry well into the output.

Yes. Veo 3.1 Fast is a speed-optimized variant that generates clips faster at lower cost per second. It's suited to rapid ideation, storyboarding, and pre-visualization where you're exploring concepts rather than generating final assets. Use full-quality Veo 3.1 for client-facing deliverables, and Veo 3.1 Fast for the exploration phases where iteration speed matters more than polished output.

Veo 3.1 is an AI video generation model from Google, available inside Masonry, the AI creative agent teams use to produce marketing, product, and brand videos.

Open a Masonry canvas, pick Veo 3.1 from the model selector, and describe the video you need: a product shot, an ad creative, a social post. Masonry generates it, then you refine, edit, and combine Veo 3.1 with other models in one workspace.

New Masonry accounts get 50 free credits to try GPT Image 2.5 Flare. Veo 3.1 is included in every paid plan, and the editor shows the exact credit cost before you generate.

Describe the scene, the motion, AND the sound you want. Veo 3.1 generates audio natively, so naming the ambience or effects gives you a more complete clip in one pass. See the prompt gallery on this page for real Veo 3.1 prompts you can copy and adapt.

Veo 3.1 is built by Google. Inside Masonry it runs alongside 50+ image and video models, so your team can pick the right one for each brief without switching tools.

Yes, the prompt gallery on this page shows real videos teams have generated with Veo 3.1 in Masonry, each paired with the exact prompt you can copy and adapt for your own brand.