TLDR
- Google introduced Flex and Priority as new Gemini API service tiers
- Flex provides 50% cost reduction for background tasks tolerant of latency
- Priority charges 75–100% premium for mission-critical, real-time operations
- Batch API continues offering 50% savings with latency up to 24 hours
- Caching tier uses token volume and retention time for pricing calculations
On April 2, Google rolled out a comprehensive pricing update for its Gemini API, introducing five separate service tiers: Standard, Flex, Priority, Batch, and Caching. This expansion provides developers with enhanced flexibility to optimize costs, performance, and dependability based on their specific application requirements.
Balance cost & reliability with our new Flex & Priority inference tiers in the Gemini API!
Flex: Pay 50% less for cost-sensitive & latency-tolerant workloads
Priority: Highest reliability for your most critical, interactive apps (with premium pricing)Together with the async… pic.twitter.com/dCCTZsQydX
— Google AI Developers (@googleaidevs) April 2, 2026
The newly introduced Flex tier targets background operations where immediate responses aren’t essential. By leveraging off-peak computational resources, it delivers 50% cost savings compared to standard rates. Response delays can span anywhere from 1 to 15 minutes without guarantees. Ideal applications include customer relationship management updates, computational research tasks, and automated agent-driven processes.
What sets Flex apart from the current Batch API is its synchronous endpoint architecture. Developers avoid the complexity of managing separate input/output files or monitoring job status queues. The interface simplifies implementation while maintaining identical cost advantages.
The Priority tier occupies the opposite position on the performance scale. With pricing 75% to 100% above standard rates, it’s engineered for time-sensitive, mission-critical applications. Latency typically ranges from milliseconds to mere seconds.
Google positions Priority for scenarios like real-time customer service chatbots, financial fraud monitoring, and automated content moderation systems. When Priority usage surpasses allocated quotas, excess requests automatically route to Standard tier instead of generating failures.
The Full Tier Breakdown
The pre-existing Batch API continues operation at 50% below standard rates, accommodating latency windows extending to 24 hours. It remains optimal for substantial offline computational tasks where timing is non-critical.
The Caching tier employs pricing models determined by token quantity and content retention duration. Google recommends this option for conversational AI with extensive system prompts, recurring analysis of large multimedia files, or searches across substantial document repositories.
Both Flex and Priority utilize the identical service_tier parameter within API calls. Developers can switch between tiers through a single configuration adjustment, with API responses confirming which tier processed each request.
Flex availability extends to all paid tier subscribers across GenerateContent and Interactions API calls. Priority access is restricted to Tier 2 and Tier 3 paid accounts using identical endpoints.
What Developers Get
The consolidated interface represents the primary innovation in this release. Previously, managing both background and interactive workflows necessitated splitting system architecture between synchronous and asynchronous frameworks. The updated system enables both workload types through identical synchronous endpoints.
Google positioned this enhancement as integral to its larger initiative supporting AI agent development, which frequently demands simultaneous handling of non-urgent background operations alongside time-critical interactive functions.
Gemini API product manager Lucia Loher and engineering lead Hussein Hassan Harrirou announced these changes on April 2, 2026.
