TPU v6e

mentions 2 type Person feed RSS

// recent coverage 2 mentions

08:18

2026-07-15

github.com

artificial-intelligence

Porting nanochat to a TPU: what carries over from PyTorch, and what breaks

A developer ported Karpathy's nanochat model to run on Google TPU v6e-8 using JAX, achieving comparable training quality but only half the hardware utilization (24% MFU vs. 47-48% on H100). The port r…

22:21

2026-06-02

letsdatascience.com

ai-infrastructure

Google Cloud Demonstrates Multi-Cluster TPU Inference Setup

Google Cloud demonstrated a multi-cluster inference setup deploying an LLM across two regional GKE clusters using TPU v6e accelerators and GKE managed DRANET for networking. The experiment, documented…

// co-occurs with top 8 entities

Google Cloud 1 Ammett Williams 1 GKE 1 Gemma 3 1 Cloud Storage 1 GKE Inference Gateway 1 Ray Serve 1 nanochat 1