Leak says Google is building a custom inference chip for Gemini
ns123abc · x · 2026-07-20
A post claims Google is developing a specialized inference chip because compute demand is so tight that Google Cloud is already turning down deals.
The leak says the new line, codenamed Frozen v2, descends from an earlier Jeff Dean design that would have etched Gemini weights directly into silicon. This version reportedly freezes the architecture instead: the Gemini architecture is hardwired into the chip, with a claimed 10x tokens-per-watt improvement over Google’s newest TPUs for inference.
Related event: Google Reportedly Developing Gemini-Specific Chip Frozen v2(10 posts)→
More from Infra
- Anthropic S-1: only $19B total compute spend since inception, overbuild warning — bookwormengr · 2026-10-27
- CoreWeave, You.com and NVIDIA lift BrowseComp 22.9% in 3-week RL post-training sprint — RichardSocher · 2026-10-27
- Grid queues make power architecture an AI-factory decision; Exowatt and Redwood offer fixes — ingliguori · 2026-10-27
- Two CVSS 9.9 Wasmtime flaws allow sandbox escape, one ARM64-only since 2020 — basedjensen · 2026-10-27
- AI Infra Meme: While Countries Need TW-Scale Power for Datacenters, People Argue About Lightbulbs — mgostIH · 2026-10-27
- NVIDIA reportedly stops taking RTX 5090 orders from board partners, even backorders — chemist_slime · 2026-10-27