AI Accelerator
An AI accelerator is hardware designed or configured to execute artificial intelligence and machine learning workloads more efficiently than a general-purpose processor executing the same workload without…
AI Compute
Explore the chips, clouds, data centers, deals and markets that supply the computing power behind AI, and what that power costs.
Start here
Understand how AI compute is measured, what it runs on, who supplies it and how it is priced. Read in order or jump to a topic.
- 1AI computeStart with what AI compute is and how it is measured
- 2Graphics processing unitSee the chip that does most of the work
- 3Cloud AI GPU Pricing ComparisonCompare what renting a GPU costs across clouds
- 4AI DatacenterFollow the buildout of AI data centers
- 5List of AI compute dealsRead the contracts that pay for the buildout
- 6Compute derivativesSee how compute is starting to trade like a commodity
Most linked
The AI Compute articles that other AI Wiki pages link to most.
- NvidiaNvidia Corporation is an American technology company that designs graphics processing units (GPUs) and is the world's dominant supplier of…
- Data CenterA data center is a purpose-built facility, or a dedicated part of a facility, that houses and interconnects information technology and…
- Google CloudGoogle Cloud is the enterprise cloud business of Google and the name of a reportable segment in Alphabet's financial statements.
- AI InfrastructureAI infrastructure is the hardware, facilities, networking, storage, and software used to develop, train, evaluate, deploy, and operate…
- IntelIntel Corporation is an American multinational semiconductor company headquartered in Santa Clara, California, that designs CPUs, GPUs, and…
- NVIDIA H100NVIDIA H100 (also called the H100 Tensor Core GPU) is a data-center graphics processing unit built by NVIDIA on the Hopper…
- Graphics processing unitA graphics processing unit (GPU) is a processor designed to execute many similar operations in parallel.
- Amazon Web ServicesAmazon Web Services (AWS) is the cloud-computing business of Amazon. It supplies computing, storage, database, networking, analytics…
Explore articles
Browse subtopics (30)
Articles that also belong to these categories. Counts cover all of AI Compute.
All articles (248)AI Hardware (179)AI Infrastructure (136)Data Centers (63)AI Companies (55)NVIDIA (55)AI Inference (25)Chinese AI (16)Google (13)AI Policy & Regulation (11)Meta AI (10)Machine Learning (8)OpenAI (7)Artificial Intelligence (6)Finance AI (5)Microsoft (5)AI Energy (4)Deep Learning (4)Training & Optimization (4)AI Benchmarks (3)Anthropic (2)Model Evaluation (2)Open Source AI (2)AI Agents (1)AI Research (1)AI Safety (1)AI Tools & Products (1)Edge computing (1)Enterprise AI (1)Research Organizations (1)Venture Capital (1)
Showing 1-60 of 248 articles
AI Accelerator Comparison (H100 vs B200 vs MI300 vs TPU)
As of July 2026, the best AI accelerator depends on the metric you care about, so the honest answer to "H100 vs B200 vs MI300 vs TPU" is not a single winner.
AI Chip
An AI chip is an integrated circuit, or a tightly integrated multi-die semiconductor package, designed or selected to execute artificial intelligence workloads efficiently.
AI Datacenter
An AI datacenter (also written as AI data center) is a facility designed to run artificial intelligence workloads on dense groups of GPUs or other accelerators.
AI Infrastructure
AI infrastructure is the hardware, facilities, networking, storage, and software used to develop, train, evaluate, deploy, and operate artificial intelligence systems. It includes more than accelerator chips.
AI OVERWATCH Act
The AI OVERWATCH Act is a United States bill introduced in the 119th Congress that would require licenses and congressional review for exports of the most advanced artificial intelligence chips
AI Task Cost
AI task cost measures the spending associated with executing a defined task using an AI system. It can be expressed per attempt, completed workflow or accepted output.
AI bubble
The AI bubble is the contested debate, most intense across 2025 and into 2026, over whether the surge in artificial intelligence investment, company valuations, and infrastructure spending that began around…
AI capital expenditure
AI capital expenditure, usually shortened to AI capex, is spending on the physical assets used to train and run artificial intelligence models: AI data centers and the land under them, servers built around…
AI chip export controls
AI chip export controls are a set of United States regulations that restrict the export, reexport, and in-country transfer of advanced artificial intelligence accelerators (graphics processing units, tensor…
AI compute
AI compute is the processing capacity used to train and run artificial intelligence models, together with the chips, memory, networks, data centers, electricity and capital that supply it.
AI data center moratorium wave (2026)
The AI data center moratorium wave of 2026 is a surge of local and state pauses, bans, and restrictive legislation aimed at hyperscale data center development, centered on the United States.
AI energy consumption
AI energy consumption is the electricity, and the associated water, land, and emissions, required to train and operate artificial intelligence systems, principally generative AI and large language models…
AMD
Advanced Micro Devices, Inc. (AMD) is an American semiconductor company that designs the CPUs, GPUs, and data-center accelerators powering one of the largest AI compute build-outs in the industry.
AMD Advancing AI 2026
Advancing AI 2026 (styled AAI 2026 in the company's own materials) was an annual product and strategy conference held by AMD in San Francisco on July 22 and 23, 2026.
AMD Helios
AMD Helios is a rack-scale artificial intelligence system from AMD that packages 72 Instinct MI455X GPUs and 18 sixth-generation EPYC "Venice" CPUs into a single double-wide cabinet
AMD Instinct MI300X
The AMD Instinct MI300X is a data center GPU accelerator that Advanced Micro Devices released on December 6, 2023, built on the CDNA 3 architecture and paired with 192 GB of HBM3 memory at 5.3 TB/s of bandwidth
AMD Instinct MI325X
The AMD Instinct MI325X is a data center GPU accelerator from AMD for AI training and inference, built on the CDNA 3 architecture with 256 GB of HBM3E memory and 6 TB/s of memory bandwidth
AMD Instinct MI350P
The AMD Instinct MI350P is a PCIe add-in card built on AMD's CDNA 4 architecture, introduced on May 7, 2026 as the third member of the Instinct MI350 series and the first AMD Instinct product in a conventional…
AMD Instinct MI355X
The AMD Instinct MI355X is a data center GPU accelerator built on AMD's CDNA 4 architecture, announced at the AMD Advancing AI 2025 event on June 12, 2025, and reaching general availability in October 2025.
AMD Instinct MI400
AMD Instinct MI400 is AMD's 2026 family of data center GPU accelerators for artificial intelligence training and inference, built on the CDNA 5 architecture.
AMD Instinct MI430X
The AMD Instinct MI430X is a data center GPU accelerator built for scientific computing and sovereign AI, and the high-precision member of the AMD Instinct MI400 series.
AMD Instinct MI455X
The AMD Instinct MI455X is a data center GPU accelerator announced by AMD on 23 July 2026 at Advancing AI 2026 in San Francisco.
ASIC
An application-specific integrated circuit (ASIC) is a chip designed to perform one job, in contrast to a general-purpose CPU or a broadly programmable GPU.
AWS Inferentia
AWS Inferentia is a family of custom application specific integrated circuits (ASICs) designed by Amazon Web Services for machine learning inference in the cloud, built to deliver, in AWS's words, "high…
AWS Trainium
AWS Trainium is a family of custom machine learning accelerator chips designed by Annapurna Labs for Amazon Web Services, purpose-built for training and, increasingly, for serving large neural networks.
AWS Trainium 2
AWS Trainium 2 (also written as Trainium2 and abbreviated Trn2) is the second generation of Amazon Web Services' custom machine learning training accelerator
AWS Trainium 3
AWS Trainium 3 (also written as Trainium3 and abbreviated Trn3) is the third-generation custom AI training and inference accelerator from Amazon Web Services, designed by Amazon's in-house chip team Annapurna…
Abilene data center (Stargate)
The Abilene data center, also called the Crusoe Abilene Stargate Campus, is the flagship artificial intelligence data center of the Stargate Project, built by Crusoe Energy on Lancium's Clean Campus outside…
Ada Lovelace (microarchitecture)
Ada Lovelace is the graphics processing unit (GPU) microarchitecture that Nvidia announced on September 20, 2022, succeeding the consumer side of the Ampere architecture and serving as the foundation for the…
Amazon Web Services
Amazon Web Services (AWS) is the cloud-computing business of Amazon. It supplies computing, storage, database, networking, analytics, security, and application services from infrastructure operated by Amazon.
Ampere (microarchitecture)
Ampere is a graphics processing unit (GPU) microarchitecture from Nvidia, announced on May 14, 2020, as the successor to the Volta and Turing architectures.
Anthropic-Amazon Trainium expansion
The Anthropic-Amazon Trainium expansion is an enlarged compute and investment agreement between Anthropic and Amazon, announced on 20 April 2026, under which Amazon committed to invest up to $25 billion more…
Anthropic-Google TPU deal
The Anthropic-Google TPU deal is a set of escalating compute and investment agreements between the AI lab Anthropic and Google, built around Anthropic running Claude on Google's tensor processing units (TPUs).
Applied Digital
Applied Digital Corporation (Nasdaq: APLD) is a United States designer, builder, and operator of large data centers for artificial intelligence and high-performance computing, headquartered in Dallas, Texas.
Axiom Space orbital data centers
Axiom Space's orbital data center (ODC) program is the Houston company's effort to sell computing and data storage in orbit to satellites and other spacecraft.
Biren Technology
Biren Technology (Chinese: 壁仞科技; pinyin: Birèn Keji) is a Chinese fabless semiconductor company, based in Shanghai and founded in 2019, that designs general-purpose GPUs and AI accelerators for data-center…
Blackhole (Tenstorrent)
Blackhole is the third-generation AI accelerator architecture from Tenstorrent, the Toronto and Santa Clara based fabless semiconductor company led by CEO Jim Keller.
Broadcom
Broadcom Inc. is an American semiconductor and infrastructure software company, traded on Nasdaq as AVGO, that has become the second most important supplier of AI compute after Nvidia by co-designing custom AI…
Broadcom Tomahawk 6
The Broadcom Tomahawk 6 is an Ethernet switch chip built for the networks that connect large clusters of AI accelerators.
CHIPS and Science Act
The CHIPS and Science Act is a United States federal law, enacted as Public Law 117-167 on August 9, 2022, that appropriated $52.7 billion for domestic semiconductor manufacturing, research, and workforce…
CME Compute Futures
CME Compute Futures are a planned family of financially settled futures contracts tied to Silicon Data benchmarks for hourly rentals of Nvidia H100 and B200 graphics processors.
Cambricon
Cambricon Technologies (Chinese: 寒武纪科技; stock ticker: 688256.SS) is a Chinese fabless semiconductor company that designs artificial intelligence processors, best known for its Siyuan (思元) Machine Learning Unit…
Cambricon Siyuan 590/690
The Cambricon Siyuan 590 and Siyuan 690 (Chinese: 思元590 / 思元690) are data-center artificial intelligence accelerators designed by Cambricon, a partially state-owned Chinese chip company often called "the…
Catalina (Meta AI rack)
Catalina is a high-power, liquid-cooled rack system designed by Meta for training and serving large AI models.
Cerebras Systems
Cerebras Systems is an American artificial intelligence hardware company that builds wafer-scale processors for AI training and inference
Cerebras WSE-3
The Cerebras WSE-3 (Wafer-Scale Engine 3) is the third-generation wafer-scale AI chip developed by Cerebras Systems, announced on March 13, 2024, and described by Cerebras as the largest chip ever built.
Cerebras WSE-4 and 2026 IPO
The Cerebras WSE-4 (Wafer Scale Engine 4) is the anticipated fourth generation of the dinner-plate-sized wafer-scale AI processor built by Cerebras Systems.
China's semiconductor industry
China's semiconductor industry is the network of wafer fabrication plants, equipment and materials suppliers, chip design houses, packaging and test operations, and state investment vehicles that produce…
Cloud AI GPU Pricing Comparison
Cloud AI GPU pricing is difficult to compare from a headline hourly rate alone. Providers sell different GPU variants, node sizes, CPU and memory bundles, regions, network configurations, and purchase models.
Cloud TPU
Cloud TPU is Google Cloud's offering of Tensor Processing Units (TPUs), the family of custom application-specific integrated circuits (ASICs) that Google builds to accelerate machine learning training and…
ClusterMAX
ClusterMAX is a rating and ranking system for GPU cloud providers published by the research firm SemiAnalysis.
Co-Packaged Optics
Co-packaged optics (CPO) is a hardware architecture that places active optical engines on the same first-level package substrate as a host application-specific integrated circuit (ASIC).
Compute Scarcity
Compute scarcity is a shortage of usable computing capacity relative to demand under specified technical, commercial and delivery conditions.
Compute derivatives
Compute derivatives are financial contracts whose value depends on the price of AI computing capacity.
Compute governance
Compute governance is a policy framework that uses regulation of the computational resources used to train and run artificial intelligence systems as a primary lever for governing advanced AI.
CoreWeave
CoreWeave, Inc. is an American artificial intelligence cloud computing company headquartered in Livingston, New Jersey, that rents large clusters of NVIDIA GPUs to AI developers and enterprises and builds the…
Cowboy Space
Cowboy Space Corporation, called Aetherflux until May 2026, is a US space company based in San Carlos, California. Robinhood co-founder Baiju Bhatt founded it in 2024.
Crusoe
Crusoe is an American AI infrastructure company headquartered in Denver, Colorado, that develops and operates data centers for artificial intelligence workloads, rents GPU capacity through its Crusoe Cloud…
Data Center
A data center is a purpose-built facility, or a dedicated part of a facility, that houses and interconnects information technology and telecommunications equipment together with the power…