Google Gemini 2.5 Flash and Pro Now in GA, Paired with 2.5 Flash-Lite Debut and Gemini CLI

Google has officially released stable, generally available versions of Gemini 2.5 Flash and Pro, two AI models with radically improved reasoning and multi-step planning abilities. Google is also debuting 2.5 Flash-Lite—Google’s most cost-efficient and fastest 2.5 model yet—and Gemini CLI, an open source AI agent for developers’ terminals.

Gemini 2.5 Flash is a well-rounded model with remarkable price-performance, designed for high-throughput enterprise tasks, including large-scale summarization, responsive chat applications, and efficient data extraction. Optimized for speed, efficiency, and scale, Gemini 2.5 Flash is now available in Vertex AI.

“SmartBear uses AI to power Test Hub, its solution for building and executing regression tests for web, desktop, and mobile. With Gemini 2.5 Flash on Vertex AI, we can accelerate tasks like translating extensive manual test scripts into robust automated tests with remarkable speed and cost-effectiveness,” said Fitz Nowlan, PhD, VP of AI, SmartBear. “The ROI is multifaceted: we're empowering our customers to realize the benefits of automation execution, while simultaneously producing intent-based, resilient-to-change test plans.”

Gemini 2.5 Pro, now available in Vertex AI, is Google’s most powerful thinking model yet, forwarding maximum response accuracy paired with state-of-the-art performance. Gemini 2.5 Pro is ideal for complex coding, reasoning, and multimodal understanding use cases, such as making sense of high-volume datasets for scientific discovery or expediting the migration of critical legacy code.

“At Multimodal, we're reimagining how business and IT teams in finance and insurance co-create intelligent agentic workflows,” said Andrew McKishnie, VP of engineering, Multimodal. “Gemini 2.5’s large context window and structured reasoning unlock a level of depth and adaptability that’s been impossible before, allowing our agents to understand, reason through, and act across highly specific domain workflows. This fundamentally changes the go-to-market experience: business teams can now visualize and validate impact on day one. For industries where trust, compliance, and precision are paramount, that’s a game-changer.''

Alongside GA for Gemini 2.5 Flash and Pro, Google is unveiling Gemini 2.5 Flash-Lite in public preview, offering dramatic improvements from the 2.0 Flash-Lite iteration. As the most cost-efficient, fastest 2.5 model yet, Gemini 2.5 Flash-Lite “excels at high-volume, latency-sensitive tasks like translation and classification, with lower latency than 2.0 Flash-Lite and 2.0 Flash on a broad sample of prompts,” according to Tulsee Doshi, senior director, product management, Google Gemini.

Gemini 2.5 Flash-Lite offers the same capabilities that define Gemini 2.5, including dynamic thinking budgets, connecting to other tools such as Google Search and code execution, multimodal input, and a 1 million-token context length.

Finally, Google is introducing Gemini CLI, an open source, extensible AI agent that provides lightweight access to Gemini from directly within developers’ terminals. Enabling developers to connect to Google’s model seamlessly within their workflows, Gemini CLI is a versatile, local utility that not only excels at coding, but also enables content generation, problem solving, deep research, and task management.

“Gemini CLI provides powerful AI capabilities, from code understanding and file manipulation to command execution and dynamic troubleshooting. It offers a fundamental upgrade to your command line experience, enabling you to write code, debug issues, and streamline your workflow with natural language,” said Taylor Mullen, senior staff software engineer, and Ryan J. Salva, senior director, product management, Google.

Gemini CLI incorporates powerful model capabilities from within the command line, including:

• Grounding prompts with Google Search, delivering real-time, external context to the model
• Built-in support for the Model Context Protocol (MCP) or bundled extensions that extend Gemini CLI’s capabilities
• Customizable prompts and instructions that help configure Gemini for developers’ unique needs and workflows
• Automate tasks and integrate existing workflows via invoking Gemini CLI non-interactively within your scripts

Google emphasizes that as a fully open source (Apache 2.0) solution, developers are invited to inspect the code, verify its security implications, and contribute to the project.

To learn more about Google’s latest AI advancements, please visit https://ai.google/.