Fluid compute is now the default for new projects
Vercel just made Fluid compute the default. Up to 85% cost savings for high-concurrency AI workloads.
Why it matters
Vercel's shift to Fluid compute by default lowers the infrastructure cost barrier for deploying AI agents and LLM applications at scale, making serverless AI deployment more economically competitive with traditional cloud.
The key facts
8 to knowFluid compute now default for all new Vercel projects
Up to 85% cost reduction for high-concurrency workloads
Reuses existing instances before spawning new ones
Combines server efficiency with serverless flexibility
Full Node.js and Python support
Pay-as-you-go model with minimal cold starts
Minimal cold starts with scale-from-zero capability
Launched following general availability and adoption across large-scale production apps
Go to the source
Vercel Blogvercel.com
Publisher excerpt: New Vercel projects now run on by default.Fluid compute This update follows , its adoption across large-scale production apps, and all deployments by default.Fluid’s general availabilityv0.devshipping with Fluid enabled Fluid compute reuses existing instances before spawning new ones, cutting costs…