CUDA Runtime API
E764026
The CUDA Runtime API is a high-level programming interface that simplifies developing and managing GPU-accelerated applications on NVIDIA GPUs.
All labels observed (6)
| Label | Occurrences |
|---|---|
| CUDA Runtime API canonical | 4 |
| CUDA runtime | 3 |
| CUDA APIs | 1 |
| CUDA Runtime API Reference Manual | 1 |
| CUDA runtime API | 1 |
| CUDA runtime library | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T8823750 — 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: CUDA Runtime API Context triple: [CUDA, provides, CUDA Runtime API]
-
A.
CUDA Driver API
The CUDA Driver API is a low-level programming interface from NVIDIA that gives developers fine-grained control over GPU resources and execution for CUDA applications.
-
B.
CUDA libraries
CUDA libraries are a collection of NVIDIA-provided GPU-accelerated software libraries that offer optimized routines for tasks such as linear algebra, deep learning, signal processing, and parallel algorithms on CUDA-enabled hardware.
-
C.
NVIDIA CUDA
NVIDIA CUDA is a parallel computing platform and programming model that enables developers to use NVIDIA GPUs for general-purpose high-performance computing.
-
D.
cuDNN
cuDNN is NVIDIA’s GPU-accelerated library of optimized primitives for deep neural networks, widely used to speed up training and inference in frameworks like TensorFlow and PyTorch.
-
E.
cuBLAS
cuBLAS is NVIDIA’s GPU-accelerated implementation of the BLAS linear algebra library, providing high-performance matrix and vector operations for CUDA applications.
- 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: CUDA Runtime API Target entity description: The CUDA Runtime API is a high-level programming interface that simplifies developing and managing GPU-accelerated applications on NVIDIA GPUs.
-
A.
CUDA Driver API
The CUDA Driver API is a low-level programming interface from NVIDIA that gives developers fine-grained control over GPU resources and execution for CUDA applications.
-
B.
CUDA libraries
CUDA libraries are a collection of NVIDIA-provided GPU-accelerated software libraries that offer optimized routines for tasks such as linear algebra, deep learning, signal processing, and parallel algorithms on CUDA-enabled hardware.
-
C.
NVIDIA CUDA
NVIDIA CUDA is a parallel computing platform and programming model that enables developers to use NVIDIA GPUs for general-purpose high-performance computing.
-
D.
cuDNN
cuDNN is NVIDIA’s GPU-accelerated library of optimized primitives for deep neural networks, widely used to speed up training and inference in frameworks like TensorFlow and PyTorch.
-
E.
cuBLAS
cuBLAS is NVIDIA’s GPU-accelerated implementation of the BLAS linear algebra library, providing high-performance matrix and vector operations for CUDA applications.
- F. None of above. chosen
Statements (55)
| Predicate | Object |
|---|---|
| instanceOf |
GPU programming interface
ⓘ
application programming interface ⓘ software library ⓘ |
| abstracts | CUDA Driver API ⓘ |
| compatibleWith |
Linux operating systems
ⓘ
Windows operating systems ⓘ
linked to:
Windows
macOS on supported legacy versions ⓘ |
| developer |
NVIDIA
ⓘ
linked to:
NVIDIA Corporation
|
| documentedIn |
CUDA C Programming Guide
ⓘ
linked to:
NVIDIA CUDA Programming Guide
CUDA Runtime API Reference Manual ⓘ
linked to:
CUDA Runtime API
|
| errorTypePrefix | cudaError_t ⓘ |
| hasComponent |
device management functions
ⓘ
error handling functions ⓘ event management functions ⓘ graph execution functions ⓘ kernel launch functions ⓘ memory management functions ⓘ occupancy calculation functions ⓘ peer-to-peer memory access functions ⓘ stream management functions ⓘ texture and surface reference functions ⓘ |
| introducedBy |
CUDA 1.0
ⓘ
linked to:
NVIDIA CUDA
|
| license | proprietary license from NVIDIA ⓘ |
| operatesOn |
NVIDIA GPUs
ⓘ
linked to:
GPU
|
| partOf |
CUDA Toolkit
ⓘ
linked to:
CUDA toolkit
CUDA platform ⓘ
linked to:
NVIDIA CUDA
|
| provides |
automatic device initialization
ⓘ
higher-level interface than CUDA Driver API ⓘ implicit context management ⓘ runtime compilation support via NVRTC integration ⓘ stream-ordered memory operations ⓘ synchronization primitives ⓘ unified memory management functions ⓘ |
| requires |
CUDA-capable NVIDIA GPU
ⓘ
supported NVIDIA driver ⓘ |
| supports |
C programming language
ⓘ
C++ programming language ⓘ Fortran via CUDA Fortran bindings ⓘ Python via wrapper libraries ⓘ |
| supportsFeature |
CUDA events
ⓘ
CUDA graphs ⓘ
linked to:
CUDA libraries
CUDA streams ⓘ managed memory ⓘ multi-GPU peer access ⓘ unified virtual addressing ⓘ |
| typicalCallPrefix | cuda ⓘ |
| usedFor |
GPU-accelerated application development
ⓘ
asynchronous execution control ⓘ general-purpose computing on GPUs ⓘ heterogeneous computing ⓘ launching GPU kernels ⓘ managing GPU memory ⓘ managing NVIDIA GPU devices ⓘ multi-GPU programming ⓘ stream and event management ⓘ |
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: CUDA Runtime API Description of subject: The CUDA Runtime API is a high-level programming interface that simplifies developing and managing GPU-accelerated applications on NVIDIA GPUs.
Referenced by (11)
Full triples — surface form annotated when it differs from this entity's canonical label.
subject linked to:
Compute Unified Device Architecture
linked to: CUDA Runtime API
subject linked to:
CUDA toolkit
linked to: CUDA Runtime API
subject linked to:
CUDA toolkit
linked to: CUDA Runtime API
linked to: CUDA Runtime API