Normalized comparison

Vercel vs Cloudflare Workers Limits

Choose by workload shape, not one headline number: Vercel documents wall-clock Function duration and payload ceilings, while Workers separates active CPU time from HTTP wall time and uses an isolate memory model.

Verified Aug 22, 2026VercelCloudflare Workers
Values are comparable only within the scope shown. Runtime, plan, model, invocation mode, and account-specific capacity can change the effective result.

Quick comparison

CriterionVercelCloudflare WorkersComparability note
Primary execution limiterMaximum function duration1,800 seconds (30 minutes)Pro and EnterpriseVercelActive CPU time30 seconds default; up to 5 minutesWorkers PaidCloudflareWall-clock function duration and active CPU time are not directly comparable.
MemoryHobby function memory2 GB / 1 vCPUHobbyVercelMemory per isolate128 MBFree and PaidCloudflareA Worker isolate may handle concurrent requests; Vercel's value is function runtime memory.
Request/response bodyFunction payload4.5 MBPayloads over the documented limit return FUNCTION_PAYLOAD_TOO_LARGE with status 413.VercelDepends on Cloudflare account planWorkers docs list plan-dependent request body limits and no enforced response-body limit.Cloudflare account-plan request limits are distinct from Workers plan CPU limits.

Which is best for your requirement?

Long I/O-heavy HTTP functions

Compare Vercel Fluid compute duration with Workers' no-hard-wall-time HTTP model and your client timeout.

CPU-bound edge logic

Model active CPU needs explicitly against Workers limits.

Large payloads

Avoid proxying through a small function payload cap; evaluate direct storage patterns.

Validate the decision with your workload

Before choosing between Vercel and Cloudflare Workers, reproduce the comparison with the exact plans, models, regions, runtimes, and invocation paths you intend to operate. The reviewed rows cover primary execution limiter and memory and request/response body; they do not turn different pricing, reliability, developer experience, or ecosystem tradeoffs into one universal score.

  1. Capture representative request sizes, token usage, duration, concurrency, storage, and failure behavior at realistic percentiles.
  2. Test the boundary and the recovery path on both candidates, including throttling, timeouts, partial failure, retries, and cost controls.
  3. Record which scoped observation drove the choice and recheck its official source before migration or a major traffic increase.

Official sources

Related

Inspect the underlying limits