Higher defaults and limits for Vercel Functions running Fluid compute
Vercel just 5x'd default execution time for serverless functions. Here's why that matters for AI agents.
Why it matters
Vercel's infrastructure upgrade to Fluid compute enables longer-running AI workloads on serverless—critical for agentic applications and real-time inference that previously timed out at 60-90 seconds. The shift to usage-based billing makes sustained compute more cost-efficient.
The key facts
13 to knowDefault execution time increased from 60-90s to 300s (5 minutes) across all plans
Maximum execution time on Pro/Enterprise plans now 800s (13+ minutes), up from 90-300s
Standard instance: 1 vCPU / 2 GB (was 1 vCPU / 1.7 GB)
Performance instance: 2 vCPU / 4 GB (was 1.7 vCPU / 3 GB)
Active CPU pricing model charges based on actual compute time, not idle memory
Infrastructure update enables longer-running AI inference and agentic workloads on serverless
Hobby plan: 300s max (previously 300s)
Pro plan: 300s default, 800s maximum (previously 90s/300s)
Enterprise plan: 300s default, 800s maximum (previously 90s/300s)
Standard instance: 1 vCPU / 2 GB RAM (previously 1 vCPU / 1.7 GB)
Performance instance: 2 vCPU / 4 GB RAM (previously 1.7 vCPU / 3 GB)
Active CPU pricing model reduces costs for memory-heavy workloads
Change published June 25, 2025
Go to the source
Vercel Blogvercel.com
Publisher excerpt: The default limits for Vercel Functions using have increased, with longer execution times, more memory, and more CPU.Fluid compute The default execution time, for all projects on all plans, is now 300 seconds (5 minutes): Memory and CPU instance sizes have also been updated: These increased…