PLATFORMS Signal 94
Compute that takes any shape
↗
Vercel announced Fluid, a unified compute layer that assembles machines to fit each workload's shape, running functions, sandboxes, and builds on shared infrastructure rather than separate primitives.
Engineers deploying on Vercel no longer pick a compute primitive per workload type; the platform matches configuration to the job and bills only for active CPU time. This removes infrastructure provisioning as a bottleneck for iteration, particularly for agent workloads that need machines provisioned faster than standard VMs allow.
Written by elseif from the cluster below · every claim links back to a sourceThe three things worth knowing
Fluid runs over 15 million builds a day, 25 million sandboxes a week, and a trillion requests a month on a single system.
Hive provisions isolated VMs in milliseconds, Fluid images boot custom environments using VHS format, and Vercel Drives provide portable storage that detaches from compute.
Active CPU pricing charges only while code executes, not while it waits on databases or models.
THE CLUSTER