Gemini 2.5 Flash and Pro, Live API, and Veo 2 in the Gemini API

Updates to the Gemini API, including the production readiness of Veo 2 for video generation, the preview of the Live API for real-time interactions, and the upcoming Gemini 2.5 Flash model, alongside the existing Gemini 2.5 Pro aim to enhance developer capabilities in building AI applications with improved thinking models, dynamic interactions, and high-quality video generation.

Google logo
Google
•
Shrestha Basu Mallick, Alisa Fortin, Ivan Solovyev, Joana Carrasqueira
•4 min read•beginner•
--
•View Original
GeminiVertex AI

Overview

The article introduces significant updates to the Gemini API, including the release of Gemini 2.5 Pro, the upcoming Gemini 2.5 Flash, and the production readiness of Veo 2 for video generation. It emphasizes the enhancements in real-time interactions through the Live API and showcases new capabilities for developers in building advanced applications.

What You'll Learn

1

How to utilize Gemini 2.5 Pro for developing visually compelling web applications

2

Why Gemini 2.5 Flash is beneficial for low latency and cost-efficient applications

3

How to generate high-quality videos using Veo 2 from text and image prompts

4

When to implement the Live API for real-time interactions in applications

Key Questions Answered

What are the new features of Gemini 2.5 Pro?
Gemini 2.5 Pro introduces advanced thinking models that can reason before responding, making it the most capable AI model for creating visually compelling web apps and developing agentic programming applications. It is now available for developers through the Gemini API and Google AI Studio.
How does Veo 2 enhance video generation capabilities?
Veo 2 allows developers to generate high-quality videos from both text and image prompts, supporting Text-to-Video (t2v) and Image-to-Video (i2v) functionalities. This empowers applications to create dynamic cinematic experiences with improved realism and motion accuracy.
What improvements are included in the Live API for Gemini models?
The Live API now supports 30 new languages, configurable Voice Activity Detection (VAD), and virtually infinite sessions through a sliding context window. These features enhance the ability to build interactive applications that process streaming audio, video, and text with low latency.
What is the pricing for video generation with Veo 2?
The pricing for video generation using Veo 2 is set at $0.35 per second of video generated, with a maximum length of 8-second video clips at 720p resolution and 24 frames per second.

Key Statistics & Figures

Maximum video length
8 seconds
Veo 2 allows for the generation of video clips up to 8 seconds long.
Video resolution
720p
Veo 2 generates videos at a resolution of 720p at 24 frames per second.
Pricing for video generation
$0.35 per second
This is the cost associated with generating video content using Veo 2.

Technologies & Tools

Some links below are affiliate links. We may earn a commission if you make a purchase.

API
Gemini API
Used for accessing the capabilities of Gemini 2.5 Pro, Gemini 2.5 Flash, and Veo 2.
Development Environment
Google AI Studio
Platform for developers to build and test applications using the Gemini API.
Cloud Service
Vertex AI
Enterprise customers can access Gemini 2.5 Pro through Vertex AI.

Key Actionable Insights

1
Leverage Gemini 2.5 Pro to enhance your web applications with advanced AI capabilities.
By utilizing the reasoning capabilities of Gemini 2.5 Pro, developers can create more interactive and intelligent applications that respond effectively to user inputs.
2
Implement the Live API to build applications that require real-time interactions.
The Live API's support for dynamic, low-latency interactions makes it ideal for applications that need to process audio and video streams, enhancing user engagement.
3
Use Veo 2 for generating high-quality video content directly within your applications.
With capabilities for both text and image prompts, Veo 2 can significantly reduce production time and improve the quality of video content, making it a valuable tool for developers in creative industries.

Common Pitfalls

1
Underestimating the complexity of integrating real-time features with the Live API.
Developers may find that building applications with real-time interactions requires careful planning and understanding of the API's capabilities to avoid latency issues.