Vercel Functions can now accept 20MiB request bodies
A public beta raises the request body cap on Vercel Functions from 4.5MB to 20MiB, behind a new environment variable and a 4GB memory setting.
Vercel announced on 20 November that Functions can now accept request bodies up to 20MiB, in public beta, up from the existing 4.5MB cap. It applies to the Node.js, Python and Bun runtimes, and to Routing Middleware.
Turning it on means setting VERCEL_SUPPORT_LARGE_REQUEST_BODY=1 and giving the project the 4GB performance memory setting. The Edge runtime is not included: it keeps the 4.5MB limit and returns a 413 on anything larger, so a function that needs the new ceiling has to run on Node.js instead. Vercel’s own advice is to read the body promptly rather than hold it all in memory, to avoid backpressure.
VERCEL_SUPPORT_LARGE_REQUEST_BODY=1
Why it matters: a 4.5MB cap is tight for anything handling file uploads, large webhook payloads, or bulk API requests, and has pushed people toward direct-to-storage uploads or chunking just to get around it. A 20MiB ceiling removes that workaround for a lot of ordinary cases, as long as the function is not running on the Edge runtime.
The caveat: it is public beta, opt-in per project, and the higher memory tier costs more to run than the default.