Compute Unified Device Architecture
E209963
Compute Unified Device Architecture (CUDA) is NVIDIA's parallel computing platform and programming model that enables developers to use GPUs for general-purpose processing to accelerate computationally intensive applications.
All labels observed (1)
| Label | Occurrences |
|---|---|
| Compute Unified Device Architecture canonical | 3 |
How this entity was disambiguated
This entity first appeared as the object of triple T1893415 — 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: Compute Unified Device Architecture Context triple: [NVIDIA CUDA, marketedAs, Compute Unified Device Architecture]
-
A.
The Universal Operating System
The Universal Operating System is the official motto of the Debian project, emphasizing its goal of providing a versatile, freely available operating system that runs on a wide range of hardware and use cases.
-
B.
Independent Computing Architecture
Independent Computing Architecture (ICA) is Citrix's proprietary protocol for delivering virtual applications and desktops over a network, enabling remote access to centralized computing resources.
-
C.
Pixelworks
Pixelworks is a semiconductor company known for designing and marketing video and display processing solutions for consumer electronics and digital projection devices.
-
D.
GM2900 platform
The GM2900 platform is a General Motors front-wheel-drive car architecture used for various compact and mid-size models, including later generations of the Saab 900.
-
E.
Snapdragon system-on-chip
The Snapdragon system-on-chip is a family of mobile processors widely used in smartphones and other devices, integrating CPU, GPU, modem, and other components to deliver high performance and power efficiency.
- 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: Compute Unified Device Architecture Target entity description: Compute Unified Device Architecture (CUDA) is NVIDIA's parallel computing platform and programming model that enables developers to use GPUs for general-purpose processing to accelerate computationally intensive applications.
-
A.
The Universal Operating System
The Universal Operating System is the official motto of the Debian project, emphasizing its goal of providing a versatile, freely available operating system that runs on a wide range of hardware and use cases.
-
B.
Independent Computing Architecture
Independent Computing Architecture (ICA) is Citrix's proprietary protocol for delivering virtual applications and desktops over a network, enabling remote access to centralized computing resources.
-
C.
Pixelworks
Pixelworks is a semiconductor company known for designing and marketing video and display processing solutions for consumer electronics and digital projection devices.
-
D.
GM2900 platform
The GM2900 platform is a General Motors front-wheel-drive car architecture used for various compact and mid-size models, including later generations of the Saab 900.
-
E.
Snapdragon system-on-chip
The Snapdragon system-on-chip is a family of mobile processors widely used in smartphones and other devices, integrating CPU, GPU, modem, and other components to deliver high performance and power efficiency.
- F. None of above. chosen
Statements (58)
| Predicate | Object |
|---|---|
| instanceOf |
GPGPU framework
ⓘ
parallel computing platform ⓘ programming model ⓘ |
| abbreviation |
CUDA
ⓘ
linked to:
NVIDIA CUDA
|
| architectureType | SIMT ⓘ |
| designedFor |
data analytics
ⓘ
deep learning workloads ⓘ high-performance computing ⓘ image processing ⓘ machine learning workloads ⓘ numerical simulation ⓘ parallel computing ⓘ scientific computing ⓘ video processing ⓘ |
| developedBy |
NVIDIA
ⓘ
linked to:
NVIDIA Corporation
|
| developer | NVIDIA Corporation ⓘ |
| enables | general-purpose computing on GPUs ⓘ |
| firstReleasedBy |
NVIDIA
ⓘ
linked to:
NVIDIA Corporation
|
| fullForm | Compute Unified Device Architecture ⓘ |
| includesLibrary |
NCCL
ⓘ
Thrust ⓘ cuBLAS ⓘ cuDNN ⓘ cuFFT ⓘ cuRAND ⓘ cuSPARSE ⓘ |
| provides |
CUDA Driver API
ⓘ
CUDA Runtime API ⓘ CUDA compiler nvcc ⓘ
linked to:
NVIDIA CUDA
CUDA libraries ⓘ |
| runsOn |
NVIDIA GPUs
ⓘ
linked to:
Nvidia Maxwell GPU
|
| supports | GPU acceleration of applications ⓘ |
| supportsFeature |
GPUDirect
ⓘ
dynamic parallelism ⓘ events ⓘ streams ⓘ tensor cores ⓘ unified memory ⓘ |
| supportsLanguage |
CUDA C
ⓘ
linked to:
NVIDIA CUDA
CUDA C++ ⓘ
linked to:
NVIDIA CUDA
CUDA Fortran ⓘ Java via bindings ⓘ MATLAB via bindings ⓘ OpenCL interoperability ⓘ Python via bindings ⓘ |
| targetHardware |
NVIDIA GeForce GPUs
ⓘ
linked to:
NVIDIA GeForce GPU line
NVIDIA Quadro GPUs ⓘ
linked to:
NVIDIA Quadro GPU line
NVIDIA Tesla GPUs ⓘ
linked to:
NVIDIA Tesla data center GPUs
NVIDIA data center GPUs ⓘ
linked to:
NVIDIA Tesla data center GPUs
|
| usedIn |
AI inference systems
ⓘ
AI training systems ⓘ HPC clusters ⓘ computational chemistry applications ⓘ computational finance applications ⓘ computer vision applications ⓘ medical imaging applications ⓘ scientific research applications ⓘ supercomputers ⓘ |
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: Compute Unified Device Architecture Description of subject: Compute Unified Device Architecture (CUDA) is NVIDIA's parallel computing platform and programming model that enables developers to use GPUs for general-purpose processing to accelerate computationally intensive applications.
Referenced by (3)
Full triples — surface form annotated when it differs from this entity's canonical label.