# Baseten cuts delta weight sync to 15 seconds with a 0.8-second pause

> Original title: The fastest delta weight syncs in the world

- Source: Baseten Blog
- Published: 2026-10-30T19:16:58.000Z
- Bad Influence: https://badinfluence.news/items/hqe0y7cmvb90c6f2czo7m6h3q
- Original: https://www.baseten.co/blog/the-fastest-delta-weight-syncs-in-the-world/

## Summary

Baseten says it rebuilt its delta weight sync path so each GPU's existing weights serve as the parent, removing a full CPU copy of the checkpoint. On GLM-5.3, a sync that took 36 seconds with a 6-second pause now takes 15 seconds with a 0.8-second pause. The freed host memory is used for KV cache offloading, which Baseten says raises Kimi K3 cache hit rates from 2% to 92% at concurrency 48.

## Why it matters

The post explains how keeping weights on the GPU cut delta sync time and pause length, a reusable pattern for RL rollout engineering.