TPUs

Tensor Processing Units designed specifically for machine learning matrix operations and neural network training.

5 companies

Google DeepMind TPU

US · Mountain View, CA

Trillium TPU v6 (2024). 4.7x improvement over v5, custom Axion ARM CPUs.

TPUs AI Accelerators

— View details →

Cerebras Systems

US · Sunnyvale, CA

Wafer-Scale Engine (WSE-3): 900K cores, 4T transistors. OpenAI $20B+ chip supply deal. $1B Series H (Feb 2026) at $23B valuation. Filed Nasdaq IPO April 2026 targeting $35B+ valuation.

TPUs AI Accelerators

$23B valuation (IPO filed) View details →

SambaNova

US · Palo Alto, CA

Reconfigurable Dataflow Units (RDU). Intel acquisition rejected; $350M strategic investment from Intel Capital (Feb 2026) + partnership. SambaNova Cloud expanding on Intel Xeon infrastructure. Enterprise AI inference focus.

AI Accelerators TPUs

$350M (Intel, Feb 2026) View details →

AWS (Trainium/Inferentia)

US · Seattle, WA

Trainium 2 and Inferentia 2. Anthropic training on 500K Trainium2 chips.

TPUs Custom Silicon

— View details →

Google (TPU)

US · Mountain View, CA

TPU v7 Ironwood (GA April 2025): 4,614 TFLOPS FP8, 192GB HBM, scales to 9,216 chips / 42.5 exaflops. Inference-first architecture. Trillium (v6) remains deployed at scale.

TPUs Custom Silicon

— View details →

Explore Other Technologies

Quantum Computing GPUs Photonic Computing Neuromorphic Computing ASICs FPGAs Analog Computing In-Memory Computing