app.title

AI server chips

Technology

Custom silicon chips that Apple is reportedly developing for internal use to run AI inference tasks, reinforcing the company's long-standing strategy of vertical integration.

entitydetail.created_at

8/19/2025, 9:47:18 PM

entitydetail.last_updated

8/19/2025, 9:52:13 PM

entitydetail.research_retrieved

8/19/2025, 9:52:13 PM

Summary

AI server chips are specialized hardware components crucial for accelerating artificial intelligence workloads, enabling complex AI algorithms and functions beyond the capabilities of traditional CPUs. They encompass various types, including Graphics Processing Units (GPUs), Field-Programmable Gate Arrays (FPGAs), and Application-Specific Integrated Circuits (ASICs), designed to efficiently handle the demanding compute environments of AI tasks. Key players in this market include NVIDIA, AMD, Intel, and cloud providers like AWS and Google, who develop both general-purpose and custom AI chips. For instance, AMD's Zen 5 microarchitecture, codenamed "Nirvana," powers its Epyc 9005 server processors, "Turin," while Apple leverages a vertical integration strategy to produce its own custom AI server chips. These chips are vital for enhancing AI capabilities in diverse applications, from data centers and public cloud servers to edge computing devices.

Referenced in 1 Document
Research Data
Extracted Attributes
  • Type

    Specialized hardware components

  • Function

    Accelerate artificial intelligence workloads, enable core AI functions, process data at the edge

  • Categories

    Graphics Processing Units (GPUs), Field-Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs)

  • Applications

    Enterprise AI factories, multimodal AI inference, physical AI, digital twins (NVIDIA Omniverse), data centers, public cloud servers, edge computing, automotive, on-premises servers

  • Key Producers

    NVIDIA, AMD, Intel, AWS, Alphabet (Google), Alibaba, IBM, Huawei, Apple, Qualcomm, Tenstorrent, Mythic

  • Key Design Features

    Incorporate large numbers of smaller transistors, AI-optimized design features for parallel processing of identical calculations

  • Examples of Chips/Architectures

    AMD Zen 5 ("Nirvana"), AMD Epyc 9005 ("Turin"), NVIDIA Blackwell Ultra, Intel Gaudi 3, AWS Trainium3, Alphabet Ironwood, Alibaba ACCEL, IBM NorthPole, Huawei Ascend 920, Qualcomm Cloud AI 100, Tenstorrent Wormhole n150/n300, Mythic M1076 Analog Matrix Processors, Envise Chip, Grayskull

  • Fabrication Process (AMD Zen 5)

    TSMC N4P (current), TSMC N3E (future)

Timeline
  • AMD's Zen 5 microarchitecture, codenamed "Nirvana," is shown on their roadmap. (Source: Wikipedia)

    2022-05-01

  • AMD's Zen 5 microarchitecture is launched for mobile processors. (Source: Wikipedia)

    2024-07-01

  • AMD's Zen 5 microarchitecture is launched for desktop processors. (Source: Wikipedia)

    2024-08-01

  • NVIDIA announces RTX PRO Servers powered by RTX PRO 6000 Blackwell Server Edition GPUs at Computex. (Source: Web Search)

    2025-05-01

  • AMD acquires a talented team of AI hardware and software engineers from Untether AI to enhance its AI capabilities. (Source: Web Search)

    2025-01-01

  • AMD acquires compiler startup Brium to optimize AI performance on its Instinct data center GPUs. (Source: Web Search)

    2025-01-01

Wikipedia
View on Wikipedia
Zen 5

Zen 5 ("Nirvana") is the name for a CPU microarchitecture by AMD, shown on their roadmap in May 2022, launched for mobile in July 2024 and for desktop in August 2024. It is the successor to Zen 4 and is currently fabricated on TSMC's N4P process. Zen 5 is also planned to be fabricated on the N3E process in the future. The Zen 5 microarchitecture powers Ryzen 9000 series desktop processors (codenamed "Granite Ridge"), Epyc 9005 server processors (codenamed "Turin"), and Ryzen AI 300 thin and light mobile processors (codenamed "Strix Point").

Web Search Results
  • Top 20 AI Chip Makers: NVIDIA & Its Competitors in 2025

    Announced in May 2025, at Computex, NVIDIA introduced RTX PRO Servers powered by RTX PRO 6000 Blackwell Server Edition GPUs, designed for enterprise AI factories. These servers deliver universal acceleration for AI, design, engineering, and business applications, supporting workloads like multimodal AI inference, physical AI, and digital twins on the NVIDIA Omniverse platform. [...] | Vendor | Category | Selected AI chip | | --- | --- | --- | | NVIDIA | Leading producer | Blackwell Ultra | | AMD | Leading producer | MI400 | | Intel | Leading producer | Gaudi 3 | | AWS | Public cloud & chip producer | Trainium3 | | Alphabet | Public cloud & chip producer | Ironwood | | Alibaba | Public cloud & chip producer | ACCEL | | IBM | Public cloud & chip producer | NorthPole | | Huawei | Public cloud & chip producer | Ascend 920 | [...] In 2025, AMD announced the acquisition of a talented team of AI hardware and software engineers from Untether AI, a developer of energy-efficient AI inference chips for edge providers and enterprise data centers. This move enhances AMD’s AI compiler, kernel development, and chip design capabilities, further strengthening its position in the inference market. Additionally, AMD acquired compiler startup Brium by to optimize AI performance on its Instinct data center GPUs for enterprise

  • What is an AI chip? | IBM

    The term “AI chip” is broad and includes many kinds of chips designed for the demanding compute environments required by AI tasks. Examples of popular AI chips include graphics processing units (GPUs), field programmable gate arrays (FPGAs) and application-specific integrated circuits (ASICs). While some of these chips aren’t necessarily designed specifically for AI, they are designed for advanced applications and many of their capabilities are applicable to AI workloads. [...] But as the complexity of the problems AI tackles increases, so do demands on compute processing and speed. AI chips are designed to meet the demands of highly sophisticated AI algorithms and enable core AI functions that aren’t possible on traditional central processing units (CPUs). [...] Edge computing, a computing framework that brings enterprise applications and additional computing power closer to data sources like Internet of Things (IoT) devices and local edge servers, can use AI capabilities with AI chips and run ML tasks on edge devices. With an AI chip, AI algorithms can process data at the edge of a network, with or without an internet connection, in milliseconds. Edge AI enables data to be processed where it is generated rather than in the cloud, reducing latency and

  • AI Chips: What They Are and Why They Matter - CSET

    Like general-purpose CPUs, AI chips gain speed and efficiency (that is, they are able to complete more computations per unit of energy consumed) by incorporating huge numbers of smaller and smaller transistors, which run faster and consume less energy than larger transistors. But unlike CPUs, AI chips also have other, AI-optimized design features. These features dramatically accelerate the identical, predictable, independent calculations required by AI algorithms. They include executing a large [...] AI chips include graphics processing units (GPUs), field-programmable gate arrays (FPGAs), and application-specific integrated circuits (ASICs) that are specialized for AI. General-purpose chips like central processing units (CPUs) can also be used for some simpler AI tasks, but CPUs are becoming less and less useful as AI advances. (Section V(A).) [...] 1. Our definition of “AI chips” includes graphics processing units (GPUs), field-programmable gate arrays (FPGAs), and certain types of application-specific integrated circuits (ASICs) specialized for AI calculations. Our definition also includes a GPU, FPGA, or AI-specific ASIC implemented as a core on system-on-a-chip (SoC). AI algorithms can run on other types of chips, including general-purpose chips like central processing units (CPUs), but we focus on GPUs, FPGAs, and AI-specific ASICs

  • 10 top AI hardware and chip-making companies in 2025 - TechTarget

    Wormhole n150 and n300 are Tenstorrent's scalable GPUs. N300 nearly doubles every spec of n150. These chips are for network AI and are put into Galaxy modules and servers. Each server holds up to 32 Wormhole processors, 2,560 cores and 384 GB of GDDR6 memory. Editor's note:_This article was updated in July 2025 to reflect the AI chips and processors each company has to offer._ [...] Qualcomm's Cloud AI 100 chip beat Nvidia H100 in a series of tests. One test was to see the number of data center server queries each chip could carry out per watt. Qualcomm's Cloud AI 100 chip totaled 227 server queries per watt, while Nvidia H100 hit 108. The Cloud AI 100 chip also managed to net 3.8 queries per watt compared to Nvidia H100's 2.4 queries during object detection.

  • Top 20 artificial intelligence chips of choice - AI Accelerator Institute

    The Envise server has 16 Envise Chips in a 4-U server configuration, consuming only 3kW power. With an unprecedented performance, it can run the largest neural networks developed to date. Each Envise Chip has 500MB of SRAM for neural network execution without leaving the processor, and 400Gbps Lightmatter interconnect fabric for large-model scale-out. [...] Enabling high performance for power-efficient AI inference in both edge devices and servers, the PCIe card simplifies integration effort into platforms where there is a constraint of space. With four M1076 Mythic Analog Matrix Processors, or AMPs, it delivers up to 100 TOPSf AI performance and supports up to 300 million weights for complex AI workloads below 25W of power. [...] The chip can achieve 368 TOPS and as much as 23,345 sentence/second at the chip thermal design power set-point needed for a 75W bus-powered PCIe card, using BERT-based for the SQuAD 1.1 data set. Grayskull is ideal for public cloud servers, inference in data centers, automotive, and on-premises servers. ## 6. Mythic