Google Introduces Flexible Pricing Tiers for Gemini API

Google on Thursday unveiled new pricing and reliability options for its Gemini API, introducing tiered inference modes designed to give developers more flexibility in managing costs and performance guarantees.

The company announced two new inference tiers — “Flex” and “Priority” — that allow developers to choose between lower-cost access with variable throughput and premium access with guaranteed capacity and faster response times.

The Flex tier offers reduced pricing in exchange for best-effort serving, meaning requests may experience higher latency or be queued during periods of peak demand. The Priority tier provides dedicated throughput and consistent performance, aimed at production workloads where reliability is critical.

“Developers building with the Gemini API have told us they want more control over how they spend on inference,” Google said in a blog post announcing the changes. The tiered approach mirrors pricing models already established by competitors in the large language model API market.

The move reflects a broader industry trend toward more granular API pricing as AI providers seek to attract a wider range of customers, from individual developers running experiments to enterprises deploying models at scale. OpenAI, Anthropic and other major providers have similarly introduced variable pricing structures in recent months.

For developers running batch processing jobs, prototyping applications or handling non-time-sensitive workloads, the Flex tier represents a significant cost reduction. Production applications serving end users, by contrast, can opt for the Priority tier to ensure consistent response times.

The new tiers are available immediately across Gemini API endpoints. Google said existing API users will not experience any changes to their current service levels and can opt into the new pricing structure at their discretion.

Google has been aggressively expanding the Gemini API’s capabilities and developer tooling as competition intensifies among AI providers for developer adoption. The company’s cloud division reported strong growth in AI-related revenue in its most recent earnings, driven in part by increasing API usage.

Pricing details for both tiers are available in Google’s updated API documentation.