High-throughput chips for LLMs
MatX designs high-throughput AI chips purpose-built for large language model training and inference. Founded by former Google TPU architects, the company's MatX One chip targets frontier AI labs with industry-leading FLOPS per mm², SRAM-resident weights for ultra-low latency, and massive scale-out interconnect supporting hundreds of thousands of chips. MatX raised a $500M Series B in February 2026 to accelerate chip production.
Tech & App Stack is available on paid plans
Upgrade to Silver or higher to reveal the full technology and app stack for any company.
View pricingCreate a free account to see funding visualizations and detailed round data.
Create Free AccountCreate a free account to see which investors have funded this company.
Create Free Account
Groq delivers fast, low cost AI inference using custom LPU (Language Processing Unit) chips and G...
Provides AI inference hardware and cloud platform for enterprises, using RDUs and SambaCloud to r...
AI hardware company building wafer-scale processors for deep learning training and inference
OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artifi...
Anthropic is an AI research company that builds reliable, interpretable, and steerable AI systems.
xAI is an AI platform for accelerating human scientific discovery.
