15 categories and 79 products and companies, charted by an ai agent.
Data center GPU accelerators are specialized processors that dramatically speed up AI model training and inference workloads in cloud and on-premise environments. Used by cloud providers, enterprises, and AI research labs to power large language models and deep learning applications.
nvidia blackwell · nvidia hopper h100 · amd instinct mi300x · amd instinct mi350 · intel gaudi 3 · nvidia rubin
google tpu v5 · aws trainium2 · aws inferentia2 · microsoft maia 100 · meta mtia · apple neural engine m5
Specialized semiconductors designed to accelerate artificial intelligence workloads through optimized tensor operations and parallel processing. Used by cloud providers, AI labs, and enterprises running large language models and deep learning applications.
groq lpu · cerebras wse-3 · tenstorrent blackhole · graphcore ipu · rebellions atom
Semiconductor foundries manufacture custom chips at scale for AI companies, enabling deployment of specialized processors. Used by AI labs, cloud providers, and chip designers needing production capacity.
tsmc · samsung foundry · intel foundry services · rapidus
nvidia nvlink · infiniband (nvidia mellanox) · ultra ethernet consortium · intel gaudi hbm fabric
Edge AI chips are specialized processors designed to run artificial intelligence models directly on devices at the network edge, enabling real-time inference without cloud connectivity. Used by manufacturers of smartphones, IoT devices, autonomous vehicles, and industrial equipment.
qualcomm snapdragon 8 elite · apple m5 · nvidia jetson orin · intel core ultra (series 2) · mediatek dimensity 9400
nvidia dgx h200 · supermicro ai servers · dell poweredge xe9680 · tenstorrent galaxy server · gigabyte g593 series
Cloud AI compute platforms provide scalable GPU/TPU infrastructure and frameworks for training and deploying machine learning models. Used by data scientists, ML engineers, and enterprises building AI applications.
aws p5 instances · google cloud tpu pods · azure nd h100 v5 · oracle cloud gpu clusters · lambda labs gpu cloud · coreweave
AI compute orchestration manages the allocation, scheduling, and optimization of computing resources across distributed infrastructure for AI workloads. Used by enterprises, cloud providers, and AI teams running large-scale model training and inference.
run:ai · crusoe energy · vast.ai · voltage park
GPU accelerators are hardware processors that offload compute workloads from CPUs, used by researchers, enterprises, and cloud providers running machine learning, scientific computing, and graphics applications. They execute matrix operations and parallel tasks faster than general-purpose processors.
nvidia · amd instinct · intel gaudi · huawei ascend · moore threads · biren technology
Custom silicon designed and manufactured by hyperscalers or AI chip companies specifically for running AI workloads, used internally by cloud providers, enterprises, and research labs to run inference and training at scale.
google tpu · aws trainium · broadcom · marvell · microsoft maia · alchip
AI chip startups design and manufacture specialized processors for artificial intelligence workloads; used by cloud providers, enterprise data centers, and device makers to accelerate machine learning inference and training.
cerebras · groq · sambanova · tenstorrent · etched · graphcore
GPU cloud services provide on-demand access to graphics processors and specialized AI accelerators hosted in data centers, accessed remotely over the internet. Used by researchers, engineers, and businesses running machine learning training, inference, and high-performance compute workloads without owning hardware.
lambda · crusoe · together ai · nebius · runpod · modal
Networking and memory systems that connect AI compute nodes and manage data movement between processors, GPUs, and storage. Used by data center operators, hyperscalers, and AI system builders to move training and inference workloads across distributed hardware.
sk hynix · micron · arista · astera labs · ayar labs
Edge and device AI refers to AI inference and processing that runs directly on edge devices, embedded systems, and IoT hardware rather than in cloud data centers. Manufacturers of consumer electronics, industrial equipment, automotive systems, and smart devices use these chips to run models locally on phones, cameras, robots, and sensors.
qualcomm · apple silicon · hailo · nvidia jetson · mediatek