Google Supercharges Flow with AI Image Editing

Google's video creation tool Flow just became much more powerful. The tech giant has integrated its Gemini 2.5 Flash image model (internally called "Nano Banana") directly into the platform, letting users manipulate images with simple text commands.

How It Works

Want to remove a background? Just type "remove the background and add soft light." Need something more creative? Try "put it on the lunar surface." The system handles these requests instantly, outputting either transparent PNGs or fully composited images while preserving remarkable detail—right down to individual strands of hair.

The real magic happens when you drag these edited images into Flow's timeline. In our tests, transforming a street photo into a "person + fantasy forest" dynamic shot took just 30 seconds, with facial features and clothing textures remaining perfectly consistent throughout the 8-second clip.

Pricing and Availability

This new capability comes at an accessible price point:

• Free for Gemini users (with some limitations)
• Just $0.039 per image for standard use
• Enterprise-grade version available through Vertex AI

Every generated image automatically includes Google's SynthID invisible watermark—a crucial feature for content authenticity in today's AI landscape.

What's Next?

The company plans to release an API batch interface next month, specifically targeting high-volume creators in sectors like short-form video and e-commerce marketing. This could revolutionize how small businesses and influencers produce professional-looking promotional content.

Key Points:

• Natural language editing: Describe what you want in plain English
• Pixel-perfect precision: Maintains fine details better than most desktop software
• Instant video integration: Drag edited images straight into your timeline
• Affordable pricing: Starts at less than 4 cents per edit
• Coming soon: Batch API for large-scale content creation