a16z Podcast Google DeepMind Developers: How Nano Banana Was Made
320 snips
Oct 28, 2025 
Oliver Wang, Principal Scientist at Google DeepMind, and Nicole Brichtova, Group Product Manager, dive into the viral sensation, Nano Banana. They discuss the innovative merging of image and multimodal models, the surprising early reactions during testing, and how this technology can enhance both professional productivity and creative expression. The duo also explores the balance between user interfaces, character consistency, and the future impact of AI on artistic learning and collaboration, while addressing artist concerns and emerging applications.
AI Snips
Chapters
Transcript
Episode notes
Model Born From Two Teams
- The Nano Banana model emerged from combining DeepMind's Imagine models with Gemini's conversational editing focus.
- Oliver Wang recounts teams merging strengths to create the multimodal Nano Banana.
Zero-Shot Personalization Moment
- Nicole tested prompts like childhood fantasies and saw the model produce a true likeness from one photo zero-shot.
- That personal moment convinced many internally that the model had real emotional resonance.
AI Frees Creators For Creativity
- Models shift creators from tedious editing toward higher-value creative work.
- Nicole predicts professionals will spend far more time ideating and less on manual operations.