# Revolut says self-hosted models cut AI coding inference costs by roughly 90%

_Published Tuesday, November 3, 2026 at 5:28 PM EST · AI · Latest · Tier 2 — Notable_

Revolut says it cut AI coding inference costs by roughly 90 percent in less than a year by making self-hosted GLM models its default, The Register reported. The change favors models it hosts itself over commercial tools such as Claude Code.

## Sources

- [Techmeme](https://www.techmeme.com/261103/p36#a261103p36)

---
Canonical: https://techandbusiness.org/newswire/T1hDkZihYW0tBtN_zABbi-
Published: 2026-11-03T22:28:36.300Z
Story chronology: 2026-11-03T21:50:01.000Z
Retrieved: 2026-11-04T01:12:16.057Z
Publisher: Tech & Business (techandbusiness.org)