415.tech
AI & tech, from the frontlines of Silicon Valley
Vercel ships drag-and-drop deploys and cuts Python cold starts to 1.3s

Vercel ships drag-and-drop deploys and cuts Python cold starts to 1.3s

Vercel's weekly recap lands one real developer win among the routine ship list: build-time bytecode compilation now bundled with functions halves Python cold starts from 2.8s to 1.3s for median-sized functions. Drag-and-drop deploy from the homepage removes the Git and CLI requirement entirely, and AI Gateway's new streaming transcription makes live captioning and voice input buildable without waiting on complete audio files.

Source: community.vercel.com ↗

Post on XEmail

Python cold starts on Vercel just got 2x faster (dropping from 2.8s to 1.3s for median-sized functions) thanks to build-time bytecode compilation now included in function bundles.

Vercel

Why this matters

  • → Python functions now start 1.3s cold instead of 2.8s—critical for serverless latency.
  • → Drag-and-drop deploy removes Git/CLI friction entirely.
  • → Streaming transcription unlocks real-time voice apps without buffering whole files.
Python's serverless speed doubles