Google · TPU · Layer 2
Google TPU v6e (Trillium)
The previous TPU generation, still widely provisioned on Google Cloud and cheaper per hour than Ironwood.
cloud only Cannot be bought. Exists only inside one cloud, so choosing it is choosing that cloud.
Specifications
- Memory
- Not published in directly comparable terms
- Bandwidth
- Not published
- Compute
- Not published in comparable FP8 terms
- Power
- Not published
- Interconnect
- ICI
- Software
- JAX, XLA, PyTorch/XLA
- Form factor
- accelerator
- Workload
- both
Verified 2026-09-06. Fields reading “not published” are exactly that — we do not estimate a figure a vendor withholds.
The catch
Same lock-in as any TPU: Google Cloud only. Google publishes far less per-chip detail for v6e than for v7, which is itself a signal about how they want it compared.
The economics
Priced as a cloud SKU. Compare against v7 on tokens delivered per dollar, not on any spec sheet.
Cost per million tokens is the metric that decides most real purchases, and there is no neutral benchmark for it — every published figure comes from a company selling one side of the comparison. Read all of them, ours included, as claims rather than measurements.
The power question underneath this
A rack of frontier accelerators draws 120–200 kW against a 2026 average of about 27 kW, and the US grid interconnection queue exceeds 2,600 GW with waits approaching five years. Whether you can energise this part is now a harder question than whether you can buy it.
Layer 1 — Energy · The interconnection queue · Rack power density · Tokens per watt
Compared against
- Google TPU v6e (Trillium) vs Google TPU v7 (Ironwood) — One generation apart, and Google publishes far more detail about the newer one — which is itself a signal about how they want them compared.Compare with Google TPU v7 (Ironwood)
Related silicon
- Google TPU v7 (Ironwood) — 192 GB HBM3E per chip, cloud only
- AWS Trainium 2 — Not published in directly comparable terms, cloud only
- AWS Trainium 3 — 144 GB HBM3e per chip, cloud only
- Microsoft Maia 200 — Not published, cloud only
- AMD Instinct MI300X — 192 GB HBM3, rent
- AMD Instinct MI350X — 288 GB HBM3E, rent