Vercel Functions Can Now Run for Up to 30 Minutes

The execution time limit for Vercel Functions has been expanded from the previous 800 seconds to 30 minutes. This is more than a 2x increase over before, and it is being rolled out first to Pro and Enterprise teams, with plans to extend it to other runtimes in the future.

This update enables the following high-load tasks:

• Long-running LLM reasoning and tool calls
• AI response streaming that lasts for several minutes
• Document and media processing, OCR, and data extraction
• Web scraping and browser automation
• Complex Workflow steps or Queue handlers

For cost efficiency, Fluid Compute technology is applied. CPU billing occurs only while code is actually executing, and billing pauses during I/O wait times such as AI model calls or database queries.

Users can enable this feature via the maxDuration setting. For Next.js App Router, this can be set in the route file, while for other frameworks it can be configured for specific function paths via vercel.json. However, execution times exceeding 800 seconds are currently in beta and require Fluid Compute to be enabled.