Google expands its AI video toolkit with Veo 3.1
Google has rolled out updated video models—Veo 3.1 Standard and Veo 3.1 Fast—through its Gemini API, building on the earlier Veo 3 release. The update promises more realistic video output, tighter prompt adherence, and richer native audio, all designed to give filmmakers greater control over storytelling and cinematic style. Available via Google AI Studio, Vertex AI, and through the Gemini app and Flow, Veo 3.1 marks a notable step in AI-assisted video generation.
What’s new in Veo 3.1
Compared with Veo 3, the 3.1 update emphasizes more faithful execution of prompts and enhanced audio capabilities. Google says the new models can generate “richer native audio,” including natural conversational tones and synchronized sound effects, while also offering better control over narrative structure and cinematic styles. In practice, creators should expect videos that align more closely with their written prompts and visual references, with audio elements that feel native to the scene.
Key features
- <strongExtend: Continue generating a clip from an existing video, enabling seamless series or sequenced storytelling.
- Frames to Video: Start with defined first and last frames to guide overall composition and pacing.
- Ingredients to Video: Shape the look and feel using up to three reference images.
- Audio integration: All three features now include audio, adding depth and realism to generated scenes.
Pricing remains consistent with Veo 3, at $0.40 per second for the Standard tier and $0.15 per second for the Fast tier, letting teams scale experiments without immediate cost barriers.
Flow: editing and refining in a collaborative AI workflow
Beyond generation, Google’s Flow platform continues to serve as an editing hub for AI-created footage. If the first take doesn’t hit the mark, creators can iterate using Flow’s editing tools—streamlining the production loop without switching tools or export pipelines. Notable Flow capabilities include:
- Insert tool: Add new elements to any scene, with the system adjusting shadows and lighting so additions feel natural within the scene.
- Content refinement: A forthcoming feature will enable the removal of unwanted objects or characters, with automatic background reconstruction to maintain seamless continuity.
Implications for filmmakers
Veo 3.1’s improved realism and richer audio deliver a more convincing virtual production experience, reducing the gap between concept and execution. The ability to guide a video’s look via reference images (Ingredients to Video) and frame-based prompts (Frames to Video) empowers directors to experiment with mood, color palettes, and camera language at a fraction of traditional production time and cost. For content creators, marketers, and short-form producers, Veo 3.1 offers a scalable way to prototype storytelling ideas, generate scratch footage, or craft fully produced clips with nuanced audio.
Availability and access
Veo 3.1 is accessible through the Gemini API, Google AI Studio, Vertex AI, the Gemini app, and Flow, providing multiple entry points for different production environments. This omnichannel access is designed to fit a variety of workflows—from solo creators to AI-assisted studio teams—while preserving pricing parity with prior Veo releases.
Conclusion
With Veo 3.1, Google reinforces its commitment to AI-enhanced video production, combining improved prompt adherence, richer native audio, and more flexible editing within Flow. As AI tools become more embedded in filmmaking, Veo 3.1 and Flow could become standard components in modern digital storytelling, helping creators realize ideas faster without sacrificing quality.
