Executive Summary
Google has announced the immediate availability of two new models, Gemini 3.6 Flash and Gemini 3.5 Flash-Lite, on its AI Gateway platform. Gemini 3.6 Flash offers improved quality and efficiency for coding and agentic tasks, while Gemini 3.5 Flash-Lite is optimized for use as a subagent in larger workflows. Developers can access these models through the AI Gateway's unified API, which provides tools for usage tracking, cost management, and performance optimization.
Key Takeaways
* Gemini 3.6 Flash: This model enhances quality for coding, agentic tasks, and web development while consuming fewer tokens and requiring fewer model calls.
* Gemini 3.5 Flash-Lite: This lightweight version upgrades the agentic capabilities of the Flash-Lite tier, making it ideal for subagents handling specific parts of a larger task.
* Availability: Both models are available now and can be accessed by setting the model name to `google/gemini-3.6-flash` or `google/gemini-3.5-flash-lite` in the AI SDK.
* Platform: The models are integrated into AI Gateway, which offers a unified API, usage/cost tracking, retries, failover, custom reporting, and API key budgets.
* Pricing: AI Gateway reflects provider pricing directly with no markup or platform fees on inference.
Strategic Importance
This update makes Google's latest, more efficient models accessible to developers through a managed platform, strengthening AI Gateway's value proposition as a central hub for building and managing AI applications.