Google has unveiled Veo 3.1, an advanced update to its AI-powered video generation model that delivers richer native audio, improved prompt adherence, and greater control over narrative styles. Available via Gemini API, Flow, and Gemini app, Veo 3.1 empowers creators with enhanced video realism and creative flexibility.
On October 14, 2025, Google announced the launch of Veo 3.1 and Veo 3.1 Fast, updated video generation models integrated into its AI toolkit, including Gemini API, Vertex AI, the Gemini app, and Flow video editor. Building on the foundation set by Veo 3 released earlier this year, Veo 3.1 focuses on delivering more authentic videos with synchronized audio, better cinematic storytelling, and superior image-to-video conversion.
Key advancements include the introduction of rich native audio generation—spanning natural conversations to synchronized sound effects—which significantly enhances the liveliness and immersion in AI-generated clips. Moreover, Veo 3.1 offers greater narrative control through an improved understanding of cinematic styles and structure.
Google also rolled out powerful new creative features within Flow, where users can now:
Generate videos by specifying first and last frames, with Veo 3.1 filling the motion in between, complete with audio.
Use up to three reference images to maintain visual and auditory consistency of characters, objects, or settings across clips.
Insert new objects into existing footage seamlessly matched to scene style, with upcoming support for object removal.
Veo 3.1 features two pricing tiers—Standard at $0.40 per second and Fast at $0.15 per second—offering options for creators depending on speed and resolution needs. Since Flow's May launch, over 275 million videos have been created leveraging Veo technology, highlighting its impact on AI-driven video content creation.
Notable Updates:
Veo 3.1 enriches generated videos with synchronized, realistic audio including dialogues and sound effects.
The model exhibits stronger prompt adherence, accurately reflecting user inputs in video outputs.
Frame-to-Video feature allows seamless video generation from static start and end frames, complete with audio.
Ingredient-to-Video functionality supports up to three reference images to maintain visual and audio consistency.
Flow video editor integrates advanced editing tools to add or soon remove scene elements naturally.
Models accessible via Google AI Studio, Vertex AI, Gemini app, and Flow with Google AI Pro subscription.
Pricing remains unchanged from Veo 3, providing creators with versatile options depending on resource needs.
Google’s Veo 3.1 represents a significant leap towards more creative, realistic, and user-controlled AI video creation, paving the way for broader adoption in filmmaking, media production, and content creation industries.
Sources: Google official blog, TechCrunch, Economic Times, Mashable, Gadgets 360