AMD Instinct MI300 series

E1040243

GPU accelerator family data center accelerator series

The AMD Instinct MI300 series is a line of high-performance data center accelerators designed for AI, HPC, and advanced compute workloads.

Try in SPARQL Jump to: Surface forms Disambiguation Statements Elicitation Referenced by

All labels observed (5)

Label	Occurrences
AMD Instinct MI300 series canonical	4
AMD Instinct MI300A	4
AMD Instinct MI300X	4
AMD Instinct	1
AMD Instinct MI300 series accelerators	1

How this entity was disambiguated

This entity first appeared as the object of triple T13320118 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.

NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07

Target entity: AMD Instinct MI300 series
Context triple: [AMD Instinct, includesModel, AMD Instinct MI300 series]

A. AMD Instinct MI100
AMD Instinct MI100 is a high-performance data center GPU accelerator from AMD designed primarily for high-performance computing (HPC) and AI workloads.
B. AMD Instinct MI250X
The AMD Instinct MI250X is a high-performance data center GPU accelerator designed for exascale supercomputing, AI, and HPC workloads.
C. NVIDIA A100
The NVIDIA A100 is a high-performance data center GPU designed for AI, high-performance computing, and data analytics workloads, featuring advanced Tensor Core acceleration.
D. Radeon Pro Vega 20
Radeon Pro Vega 20 is a professional-grade mobile graphics processor from AMD’s Vega lineup, designed to deliver high-performance graphics and compute capabilities in thin-and-light laptops.
E. NVIDIA RTX A6000
The NVIDIA RTX A6000 is a high-end professional graphics card designed for demanding workloads like AI, data science, and advanced 3D content creation.
F. None of above. chosen
G. Unsure - the case is ambiguous/there is not enough information to decide.

NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07

Target entity: AMD Instinct MI300 series
Target entity description: The AMD Instinct MI300 series is a line of high-performance data center accelerators designed for AI, HPC, and advanced compute workloads.

A. AMD Instinct MI100
AMD Instinct MI100 is a high-performance data center GPU accelerator from AMD designed primarily for high-performance computing (HPC) and AI workloads.
B. AMD Instinct MI250X
The AMD Instinct MI250X is a high-performance data center GPU accelerator designed for exascale supercomputing, AI, and HPC workloads.
C. NVIDIA A100
The NVIDIA A100 is a high-performance data center GPU designed for AI, high-performance computing, and data analytics workloads, featuring advanced Tensor Core acceleration.
D. Radeon Pro Vega 20
Radeon Pro Vega 20 is a professional-grade mobile graphics processor from AMD’s Vega lineup, designed to deliver high-performance graphics and compute capabilities in thin-and-light laptops.
E. NVIDIA RTX A6000
The NVIDIA RTX A6000 is a high-end professional graphics card designed for demanding workloads like AI, data science, and advanced 3D content creation.
F. None of above. chosen

Statements (47)

Predicate	Object
instanceOf	GPU accelerator family ⓘ data center accelerator series ⓘ
announced	2023 ⓘ
announcedBy	AMD ⓘ linked to: Advanced Micro Devices
architecture	CDNA 3 ⓘ
brand	AMD Instinct ⓘ linked to: AMD Instinct MI300 series
category	AI accelerator ⓘ HPC accelerator ⓘ
designedFor	artificial intelligence workloads ⓘ data center workloads ⓘ high performance computing ⓘ inference workloads ⓘ large language model training ⓘ
includesModel	AMD Instinct MI300A ⓘ linked to: AMD Instinct MI300 series AMD Instinct MI300X ⓘ linked to: AMD Instinct MI300 series AMD Instinct MI325X ⓘ AMD Instinct MI350X ⓘ
manufacturer	Advanced Micro Devices ⓘ
memoryType	HBM3 ⓘ
optimizedFor	energy efficiency per FLOP ⓘ large model memory capacity ⓘ
packageType	multi‑chip module ⓘ
predecessor	AMD Instinct MI200 series ⓘ
segment	data center GPUs ⓘ
supports	BF16 precision ⓘ C++ ⓘ FP16 precision ⓘ FP8 precision ⓘ Fortran ⓘ HIP programming model ⓘ INT8 precision ⓘ Infinity Fabric interconnect ⓘ linked to: Infinity Fabric Link OpenCL ⓘ OpenMP offload ⓘ linked to: OpenMP PCI Express ⓘ high‑bandwidth memory ⓘ mixed‑precision compute ⓘ
supportsSoftware	PyTorch ⓘ ROCm ⓘ TensorFlow ⓘ
targetMarket	enterprise data centers ⓘ hyperscale data centers ⓘ
useCase	AI inference ⓘ AI training ⓘ HPC simulation ⓘ cloud computing ⓘ scientific computing ⓘ

How these facts were elicited

Referenced by (14)

Full triples — surface form annotated when it differs from this entity's canonical label.

AMD Instinct → includesModel → AMD Instinct MI300 series ⓘ

subject linked to: Instinct

AMD Instinct → includesModel → AMD Instinct MI300A ⓘ

subject linked to: Instinct