MatX One
Provider 
Founded 2022
Sells To Technology Companies
Pricing Model Pre-Revenue
Ownership Private
MatX One
ProductOperating Status: Currently Offered
MatX One is a high-throughput chip for large language model training and inference.
MatX One is a high-throughput chip for large language model training and inference.
MatX One is an accelerator chip for large language models, built on a splittable systolic array that combines SRAM-first low latency with HBM long-context support. It targets training, reinforcement learning, and inference prefill and decode for large mixture-of-experts and dense models.
The chip was announced in February 2026 and had not yet taped out as of that announcement. MatX states it delivers higher throughput than any announced product while matching the best latencies.
Advanced Micro Devices
Instinct
amd.com
NVIDIA
Nvidia H100
nvidia.com
Amazon Web Services
Trainium
aws.amazon.com
Cerebras Systems
Wafer-Scale Engine
cerebras.ai
| Attribute | MatX One | Instinct | Nvidia H100 | Trainium | Wafer-Scale Engine | |
|---|---|---|---|---|---|---|
| Provider | | | | | | |
| Founded | 2022 | 1969 | 1993 | 2006 | 2016 | |
| Sells To | Technology Companies | Consumers, Enterprise | Consumers, Enterprise | Small Business | — | |
| Pricing Model | Pre-Revenue | Software, Transactional | Licensing, Product Sales, Subscription | Recurring | — | |
| Ownership | Private | Public | Public | Public | — |
Provider
Founded 2022
Sells To Technology Companies
Pricing Model Pre-Revenue
Ownership Private
Provider
Founded 1969
Sells To Consumers, Enterprise
Pricing Model Software, Transactional
Ownership Public
Provider
Founded 1993
Sells To Consumers, Enterprise
Pricing Model Licensing, Product Sales, Subscription
Ownership Public
Provider
Founded 2006
Sells To Small Business
Pricing Model Recurring
Ownership Public
Provider
Founded 2016
Sells To —
Pricing Model —
Ownership —
Maverick-2 is NextSilicon's runtime-reconfigurable dataflow accelerator for high-performance computing and AI workloads.
TPU is Google's custom AI accelerator ASIC family delivered through Google Cloud TPU Pods for ML training and inference.
IBM Power Systems are enterprise servers built on POWER processors for mission-critical Linux, AIX, and IBM i workloads.
Maverick-2 is NextSilicon's runtime-reconfigurable dataflow accelerator for high-performance computing and AI workloads.
TPU is Google's custom AI accelerator ASIC family delivered through Google Cloud TPU Pods for ML training and inference.
Intel Gaudi is a data-center AI accelerator product family.
IBM Power Systems are enterprise servers built on POWER processors for mission-critical Linux, AIX, and IBM i workloads.
NVIDIA Blackwell is a data-center GPU architecture for generative AI and accelerated computing.
NVIDIA Hopper is a data-center GPU architecture for AI and high-performance computing workloads.
Purpose-built AI supercomputers integrating GPUs, CPUs, networking, and software for enterprise training and inference.