Gemini Omni Flash & Nano Banana 2 Lite: Speed Meets Cost

Google just dropped two new models on the Gemini Enterprise Agent Platform that are all about speed and cost performance. Think of them as the capybaras of the AI swamp—fast, friendly, and efficient, leaving the caimans (expensive competitors) in the dust.

Gemini Omni Flash: Video Editing, Now Conversational

This model brings high-quality video generation and editing into your apps with natural language. Swap characters, relight scenes, or add objects—all while keeping audio and video tracks intact. It’s like having a video editor that understands your every word.

Key Features

• Conversational editing: Use plain language to edit videos.
• Multimodal input: Combine text, images, and video to guide generation.
• World knowledge: Understands physics, history, and culture for realistic scenes.
• Text sync: Renders legible text and graphics directly into video.

Priced at $0.10 per second of video output, it’s a steal for high-quality video generation. Check out the documentation and pricing.

Nano Banana 2 Lite: Images in Four Seconds

This model is the fastest and most cost-efficient in the Nano Banana family. Generate images in as little as four seconds—perfect for rapid prototyping, A/B testing, or powering social apps at scale.

Improvements Over Nano Banana

• World knowledge: Draft accurate scenes and data visualizations.
• Character consistency: Maintain identities across generations.
• Quick text: Render legible text for localized ad variations.

Both models include C2PA content credentials and SynthID watermarks by default. For high concurrency, provisioned throughput is available for Nano Banana 2 Lite now, with Gemini Omni Flash coming soon.

Ready to dive in? Try them in Agent Studio, or check out the Nano Banana 2 Lite docs and Gemini Omni Flash docs. For more on Google Cloud, see our Google Cloud review.

Original announcement published on Google Cloud.