Netlify Adds Per-Function Compute Controls for Memory, CPU, and Regional Execution.

Netlify rolled out per-function compute controls last week, giving developers granular control over how serverless functions execute. The update targets teams deploying larger apps and workloads that demand more resources, including AI inference tasks and APIs handling heavy payloads.

The new configuration options let developers set three parameters directly in code: execution region, memory allocation (1–4 GB), and CPU allocation (0.5–2 vCPU). Memory and CPU scale together automatically, so developers can specify whichever metric makes the most sense for their use case and Netlify adjusts the other proportionally.

Regional execution addresses two core needs: performance and compliance. Deploying a function closer to users or the databases and services it depends on reduces latency, particularly for latency-sensitive workloads like real-time APIs or data-heavy operations. For teams subject to data sovereignty or compliance requirements, regional controls ensure functions process data in specific geographies.

Memory and CPU controls address a different bottleneck. Default serverless functions allocate 1 GB of memory and 0.5 vCPU, sufficient for lightweight tasks but restrictive for compute-intensive jobs. AI inference, large file processing, and complex data transformations routinely exceed these limits. Netlify's new settings let developers scale up to 4 GB of memory and 2 vCPU per function, defined at the function level rather than site-wide.

This approach follows a broader industry shift toward giving developers infrastructure control without leaving the serverless model. Vercel and other platforms have adopted similar patterns, recognizing that one-size-fits-all resource limits no longer match real-world workloads.

All compute settings live in code, which means they version with the rest of the project and get validated in deploy previews before reaching production. This prevents resource misconfigurations from impacting live traffic. The configuration format integrates with Netlify's recent push toward typed function definitions, part of their Agent Experience overhaul.

Developers on credit-based plans (Pro and above) can access per-function memory and CPU controls immediately. Regional execution is available across all plans that support function configuration.

The timing aligns with Netlify's broader repositioning around AI-native development. The platform recently added support for Agent Runners and event handlers, infrastructure specifically designed for agentic workflows. Per-function compute controls extend that capability by giving AI workloads the resources they need to run efficiently at scale.

For teams already running complex serverless architectures, the update removes a longstanding friction point. Instead of architecting around fixed resource limits or migrating compute-heavy tasks to separate infrastructure, developers can now tune individual functions to match their workload. The result is fewer workarounds and more predictable performance for resource-intensive tasks.