Google has officially begun rolling out Veo 3’s Ingredients to Video inside the Gemini app — expanding one of its most advanced creative tools beyond Flow and into its mainstream AI ecosystem. This new capability allows creators to generate videos simply by uploading reference photos, giving them more control over characters, objects, world-building, and overall visual style.

The feature is currently available to Google AI Plus, Google AI Pro, and Google AI Ultra subscribers.

How the Feature Works

Launched alongside the Veo 3.1 model on October 15, Ingredients to Video lets users upload one or multiple reference images—including those created using Imagen—to guide the look and feel of video output.

By using visual references, creators can:

  • Ensure character consistency throughout scenes
  • Transfer art styles, lighting, or color palettes
  • Build custom worlds and environments
  • Reduce dependency on lengthy text prompts
  • Achieve more predictable and professional-quality results

Additionally, the Gemini app now displays which video-generation model is in use. Under Gemini Tools, users will see “Veo 3.1” when their clip is powered by the latest engine.

How It Compares to Competitors

Even with the added control, Veo 3.1 still sits behind competitors like OpenAI’s Sora in terms of creative freedom. Google’s stricter safety filters mean there are videos it will refuse to generate, while some rival platforms allow broader experimentation.

Google’s positioning remains clear:

Fast, guided, safe video creation — ideal for short-form content.

Rollout Timeline

The simplified photo-to-video feature is rolling out today, with all paid subscribers expected to receive access by next week.

It remains exclusive to Google’s AI subscription tiers and is not available on the free Gemini plan.

FAQs

1. What makes Ingredients to Video different from regular prompt-based video generation?
Instead of relying heavily on text prompts, users can upload reference images to guide the model. This results in better character consistency, stronger artistic control, and more accurate representation of what the user has in mind.

2. Do I need high-quality reference photos for best results?
While high-resolution images help achieve better precision, the model can still work with standard-quality photos. However, for detailed scenes or consistent characters, clearer images are recommended.

Conclusion

Google’s rollout of Ingredients to Video marks a significant step toward more intuitive, controlled, and visually accurate AI video creation. By allowing users to guide clips through reference images, Veo 3.1 opens doors for filmmakers, content creators, marketers, educators, and hobbyists who want simplicity without sacrificing quality.
While still more restricted than platforms like Sora, Google’s approach prioritizes safety, ease of use, and flexibility — making AI-driven video creation faster and more accessible than ever.