Fetching from the wire…
Public story · 2026-06-30 · high
It shipped the same week as full-stack hosting and a v0 redesign, aimed at Fly and Railway's turf.
Why now: The 5GB limit is already in public beta, and the pricing question at that scale is still open.
Vercel shipped an OCI-compliant container registry this week, letting Dockerfile-based servers run on Fluid compute, per Vercel's blog.
That matters for solo builders and small teams. Vercel Functions can now package up to 5GB for Node.js and Python, a 20x jump from the old 250MB cap, still in public beta.
That ceiling used to send you looking for a container host. A chunky ML dependency, a headless browser, a big native binary. Now they fit inside a single Vercel Function.
The registry didn't launch alone. Vercel Services also went GA this week, running a Next.js frontend and a FastAPI backend on shared infrastructure in one project.
v0 shipped Design Systems 2.0 the same week, generating UI from a team's own components and fonts instead of generic output.
Yes, but the old pitch was frontend and functions, leave containers to someone else. Vercel just erased that line on its own.
I'm not moving my Slack bot and dashboard off Fly. That machine works fine already. For a new full-stack project this week, though, I'd deploy it on Vercel end to end. I'd watch two things. What a 5GB function costs under real traffic, and how Fluid compute cold starts compare to a Fly machine that never sleeps.
Convenience is real. So is the bill.
Each link below shares sources, entities, or timing with this story.
Announced August 7: Hermes can use AI Gateway as its inference layer for 200+ models with no token markup and per-request dashboard visibility, and execute shell commands inside an isolated Vercel Sandbox microVM instead of on your machine, with Node.js 24/22 and Python 3.13 a...
Python functions now compile to bytecode during build rather than at first import, a 54% median cold-start reduction with zero code changes (Vercel). Python cold start has been the main reason teams reach for Node on edge functions. This doesn't close the gap but it narrows it...
FastAPI, Django and Flask apps can now set response headers or rewrite requests to internal paths through routing rules, applied without a new deployment (Vercel). That last detail is the useful one for anyone whose deploy cycle also flushes an in-memory cache. Separately, the...
At Vercel Ship 26 in London today, Guillermo Rauch launched what he's calling the Agent Stack: AI SDK, AI Gateway, Sandbox, Workflow SDK, and Chat SDK, plus a new Vercel Connect and a set of enterprise agent controls. Vercel is repositioning the entire platform around running...
Anaconda acquired Kilo Code on July 15; Kilo supplies planning, coding, and debugging agents inside VS Code, JetBrains, and the CLI. Anaconda's moat is being the default Python environment in enterprises and universities, and that moat is worth very little if the agent layer a...
Node 20 hit end of life April 30, and Vercel is surfacing affected projects now. (Vercel) Queue the runtime bump now instead of scrambling in late September. Small item, hard deadline. Also worth grabbing: Vercel Blob added consistent reads on private storage, so useCache: fal...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.