Tesla Cortex
数据中心
AI园区
运营中
Austin, Texas, US
The GPU cluster at Giga Texas training Full Self-Driving.
您正在查看一条数据中心记录：一个真实的设施——谁运营它、里面装有什么硬件、电力从何而来。
Cortex is Tesla's training cluster at its Austin factory — roughly 50,000 H100-class GPUs per company statements, cooled by a purpose-built plant and buffered by Megapacks — the compute behind FSD's vision-only neural networks after the Dojo program wound down.
规格与关键参数
每个数据均标注来源依据：厂商规格、厂商声明、公开报道、估算或官方数据。"未公开披露"即字面含义——我们从不捏造数值。
Overview
Operator
Tesla
Scale
GPUs
~50,000 H100-class (2024 company statements)Vendor claim
它如何与整个技术栈相连
本记录的关联链接一览——点击分组可直接跳转。
运营方
来源
记录审阅截至 2026-01-15 — 查看 方法论.