Google TPU IronwoodGoogle
11
Inference-first TPU pods
- Version
- 7th-generation (Ironwood)
- Cost
- from $5.4/unit
- Model
- Gemini
Broadcom custom accelerators
Broadcom · #12 most active of 18 in Chips & Compute
Co-designed XPUs behind most hyperscaler silicon.
Current version
2025
Entry cost
Enterprise supply agreements
Changes / 30d
0
The short answers
- What is the latest version of Broadcom custom accelerators?
- The current shipped version of Broadcom custom accelerators is 2025.
- How much does Broadcom custom accelerators cost?
- Broadcom custom accelerators has no published flat entry price. Enterprise supply agreements
- What AI model does Broadcom custom accelerators use?
- Broadcom custom accelerators runs primarily on powers Google TPU.
Source: Broadcom (no public source page recorded) · Reusable under CC BY 4.0 — cite Tomorrow
Capabilities
- Custom XPU design
- Tomahawk networking
- Hyperscaler programs
Pricing
Enterprise supply agreements
Entry price over time
Not enough pricing history yet — we start charting from the second observation.
Geek mode
Models under the hood
- disclosed
powers Google TPU
Provider undisclosed
Context window
—
Public API
no
Multi-model routing
no
Stack signals
- TSMC
- Ethernet fabric
Shared model stack
Other tracked products running on the same foundation models — a quick read on how much of the catalog moves when one of these models changes.
- powers Google TPUProvider undisclosed · 1 product
green disclosed · amber inferred · grey unattributed. Aliases are folded into one model; provider concentration counts only models with an established vendor.
Reported scale
Reference data. Each figure is whatever the source actually said — weekly users, downloads, revenue run-rate — with its own definition and date. These are not comparable between tools and are never used to rank anything.
No public usage figure on record for this product.
Public market scorecard
Broadcom investor relationsBroadcom trades as NASDAQ: AVGO. Figures below are as reported for Q2 FY2026 on Jun 5, 2026; market levels are the close on Aug 6, 2026 and are a reference marker, not a live quote.
Q2 FY2026 revenue
~$17B
+22% YoY
- AI semiconductor revenue
- ~$6B
- custom accelerators + networking
- Infrastructure software
- ~$6.5B
- VMware attach
Also in Chips & Compute
All Chips & ComputeCerebras WSE-3
Cerebras
9
Wafer-scale inference and training
- Version
- CS-4
- Cost
- Free trial credits available; developer tier has free promotional credits.
- Model
- Codex-Spark
Tenstorrent Blackhole
Tenstorrent
8
Open RISC-V AI hardware
- Version
- Blackhole
- Cost
- from $999/unit
- Model
- Blackhole Tensix
7
Cloud-native training silicon
- Version
- Trainium2
- Cost
- Pricing is not published on the page. Available via EC2 instances/UltraServers.
- Model
- Neuron compiler
NVIDIA Blackwell
NVIDIA
5
GB200/B200 rack-scale AI systems
- Version
- GB200 NVL72
- Cost
- Pricing is not published (enterprise/quote-only)
- Model
- Blackwell
High-memory GPU alternative
- Version
- MI355X
- Cost
- Enterprise hardware, pricing not published
- Model
- CDNA
Change history
No changes recorded yet.
Broadcom custom accelerators compared
Straight head-to-head pages against the busiest products in Chips & Compute.
How to cite this page
Free to cite and reuse under CC BY 4.0. Permalink: https://tomorrow.aliensquad.ai/tools/broadcom-tpu
APA
Tomorrow. (2026). Broadcom custom accelerators — version, pricing and model stack [Data set entry]. AlienSquad. Retrieved 2026-11-08, from https://tomorrow.aliensquad.ai/tools/broadcom-tpu
BibTeX
@misc{tomorrow-tools-broadcom-tpu,
author = {{Tomorrow}},
title = {Broadcom custom accelerators — version, pricing and model stack},
year = {2026},
publisher = {AlienSquad},
howpublished = {\url{https://tomorrow.aliensquad.ai/tools/broadcom-tpu}},
note = {Accessed: 2026-11-08}
}