Google expands TPU v8 lineup: 8tx cuts latency 2x, 8i Max doubles silicon per rack
Beth_Kindig · x · 2026-11-13
Google is expanding its TPU v8 lineup with two optimized chips:
- TPU 8tx: up to 2x lower latency versus TPU 8t, optimized for confidential computing — small-scale training or inference on frontier/proprietary models with sensitive data.
- TPU 8i Max: built for massive context windows and maximum compute density per rack, scaling compute and memory in lockstep to deliver double the silicon, double the SRAM, and double the HBM bandwidth in the same rack footprint.
The move is framed against the ongoing compute competition with NVIDIA, AMD, and Broadcom.
Related event: Google Expands TPU v8 Lineup with 8tx and 8i Max Chips(4 posts)→
More from Infra
- Nvidia pressured to boost 25% residual value backstop on planned $500B AI financing vehicle — dinabass · 2026-11-13
- A GPU deep-dive for AI engineers: token flow, memory math, quantization and multi-GPU sharding — kmeanskaran · 2026-11-13
- TSMC confirms employee data leaked via cloud provider, says own systems not breached — TinfoilTricorn · 2026-11-13
- CoreWeave Q3 revenue hits $3.66B, up 168%; signs multi-year Wayve training deal — thursdai_pod · 2026-11-13
- 67B MiniMax H3-Image runs on a 16GB MacBook Air via MLX with ~85% less peak memory — FlimsyBrief4232 · 2026-11-13
- SimpleHC matches mHC performance with only 2.4% extra training time, study finds — dots-studio · 2026-11-13