AMD Instinct MI250X
E389990
The AMD Instinct MI250X is a high-performance data center GPU accelerator designed for exascale supercomputing, AI, and HPC workloads.
All labels observed (3)
| Label | Occurrences |
|---|---|
| AMD Instinct MI250X canonical | 4 |
| AMD Instinct MI200 series | 1 |
| AMD Instinct MI250 | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T3806553 — 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 MI250X Context triple: [Frontier, gpuModel, AMD Instinct MI250X]
-
A.
Intel Arc
Intel Arc is a line of discrete graphics processing units (GPUs) developed by Intel for gaming, content creation, and high-performance graphics workloads.
-
B.
AMD EPYC
AMD EPYC is a family of high-performance server processors from AMD designed for data centers, cloud computing, and enterprise workloads.
-
C.
AMD Radeon GCN-based GPU
The AMD Radeon GCN-based GPU is a graphics processor architecture from AMD’s Graphics Core Next family, widely used in gaming consoles and PCs for efficient, parallel graphics and compute performance.
-
D.
NVIDIA DGX
NVIDIA DGX is a line of high-performance, AI-optimized computing systems designed for training and deploying large-scale machine learning and deep learning models.
-
E.
NVIDIA Studio
NVIDIA Studio is a platform and suite of tools, drivers, and optimizations designed to enhance performance and reliability for creative and content creation workflows on NVIDIA GPUs.
- 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 MI250X Target entity description: The AMD Instinct MI250X is a high-performance data center GPU accelerator designed for exascale supercomputing, AI, and HPC workloads.
-
A.
Intel Arc
Intel Arc is a line of discrete graphics processing units (GPUs) developed by Intel for gaming, content creation, and high-performance graphics workloads.
-
B.
AMD EPYC
AMD EPYC is a family of high-performance server processors from AMD designed for data centers, cloud computing, and enterprise workloads.
-
C.
AMD Radeon GCN-based GPU
The AMD Radeon GCN-based GPU is a graphics processor architecture from AMD’s Graphics Core Next family, widely used in gaming consoles and PCs for efficient, parallel graphics and compute performance.
-
D.
NVIDIA DGX
NVIDIA DGX is a line of high-performance, AI-optimized computing systems designed for training and deploying large-scale machine learning and deep learning models.
-
E.
NVIDIA Studio
NVIDIA Studio is a platform and suite of tools, drivers, and optimizations designed to enhance performance and reliability for creative and content creation workflows on NVIDIA GPUs.
- F. None of above. chosen
Statements (50)
| Predicate | Object |
|---|---|
| instanceOf |
AMD Instinct accelerator
ⓘ
GPU accelerator ⓘ data center GPU ⓘ |
| architecture |
CDNA 2
ⓘ
linked to:
CDNA
|
| bf16MatrixPeak | ~383 TFLOPS ⓘ |
| brand |
AMD Instinct
ⓘ
linked to:
AMD Instinct MI100
|
| codeName | Aldebaran ⓘ |
| dieCount | 2 ⓘ |
| doublePrecisionPerformancePeak | ~47.9 TFLOPS ⓘ |
| family |
CDNA 2
ⓘ
linked to:
CDNA
|
| formFactor | OAM ⓘ |
| fp16MatrixPeak | ~383 TFLOPS ⓘ |
| launchDate | 2021-11 ⓘ |
| manufacturer | Advanced Micro Devices ⓘ |
| memoryBandwidth | 3276 GB/s ⓘ |
| memoryBusWidth | 8192 bit ⓘ |
| memoryCapacity | 128 GB ⓘ |
| memoryType | HBM2e ⓘ |
| packageType |
MCM
ⓘ
multi-chip module ⓘ |
| processNode |
N6
ⓘ
TSMC 6 nm ⓘ |
| singlePrecisionPerformancePeak | ~95.7 TFLOPS ⓘ |
| successor | AMD Instinct MI300 series ⓘ |
| supports |
BF16
ⓘ
FP16 ⓘ FP32 ⓘ FP64 ⓘ HBM2e ECC ⓘ HIP ⓘ Infinity Fabric Link ⓘ MPI acceleration ⓘ Matrix Cores ⓘ OpenCL ⓘ OpenMP offload ⓘ PCI Express 4.0 ⓘ ROCm ⓘ |
| targetMarket |
artificial intelligence
ⓘ
data center ⓘ exascale supercomputing ⓘ high performance computing ⓘ |
| tdp | 500 W ⓘ |
| useCase |
AI inference
ⓘ
AI training ⓘ HPC simulations ⓘ climate modeling ⓘ computational fluid dynamics ⓘ scientific computing ⓘ |
| usedIn |
Frontier supercomputer
ⓘ
exascale systems ⓘ |
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
Instruction
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Input
Subject: AMD Instinct MI250X Description of subject: The AMD Instinct MI250X is a high-performance data center GPU accelerator designed for exascale supercomputing, AI, and HPC workloads.
Referenced by (6)
Full triples — surface form annotated when it differs from this entity's canonical label.