cuSPARSE
E213160
cuSPARSE is NVIDIA’s GPU-accelerated library providing high-performance sparse linear algebra routines for CUDA applications.
All labels observed (2)
| Label | Occurrences |
|---|---|
| cuSPARSE canonical | 10 |
| cuSPARSELt | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T1893380 — 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: cuSPARSE Context triple: [NVIDIA CUDA, includes, cuSPARSE]
-
A.
cuBLAS
cuBLAS is NVIDIA’s GPU-accelerated implementation of the BLAS linear algebra library, providing high-performance matrix and vector operations for CUDA applications.
-
B.
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.
-
C.
cuRAND
cuRAND is NVIDIA's GPU-accelerated random number generation library designed to efficiently produce high-quality random numbers for parallel applications using CUDA.
-
D.
CUDA Fortran
CUDA Fortran is an extension of the Fortran programming language that enables developers to write and run parallel code on NVIDIA GPUs using the CUDA architecture.
-
E.
CuPy
CuPy is an open-source array library for Python that accelerates numerical computing by providing a NumPy-compatible interface backed by GPU execution.
- 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: cuSPARSE Target entity description: cuSPARSE is NVIDIA’s GPU-accelerated library providing high-performance sparse linear algebra routines for CUDA applications.
-
A.
cuBLAS
cuBLAS is NVIDIA’s GPU-accelerated implementation of the BLAS linear algebra library, providing high-performance matrix and vector operations for CUDA applications.
-
B.
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.
-
C.
cuRAND
cuRAND is NVIDIA's GPU-accelerated random number generation library designed to efficiently produce high-quality random numbers for parallel applications using CUDA.
-
D.
CUDA Fortran
CUDA Fortran is an extension of the Fortran programming language that enables developers to write and run parallel code on NVIDIA GPUs using the CUDA architecture.
-
E.
CuPy
CuPy is an open-source array library for Python that accelerates numerical computing by providing a NumPy-compatible interface backed by GPU execution.
- F. None of above. chosen
Statements (48)
| Predicate | Object |
|---|---|
| instanceOf |
CUDA library
ⓘ
GPU-accelerated sparse linear algebra library ⓘ software library ⓘ |
| developer |
NVIDIA
ⓘ
linked to:
NVIDIA Corporation
|
| documentationURL | https://docs.nvidia.com/cuda/cusparse ⓘ |
| domain | sparse linear algebra ⓘ |
| includedIn | CUDA Toolkit installation ⓘ |
| integratesWith |
NVIDIA HPC SDK
ⓘ
Thrust ⓘ cuBLAS ⓘ cuSOLVER ⓘ |
| license | proprietary ⓘ |
| optimizedFor | high-performance sparse computations on GPUs ⓘ |
| partOf |
NVIDIA CUDA Toolkit
ⓘ
linked to:
CUDA toolkit
|
| programmingModel |
CUDA
ⓘ
linked to:
NVIDIA CUDA
|
| providesAPI | C API ⓘ |
| requires |
CUDA runtime
ⓘ
linked to:
CUDA Runtime API
CUDA-capable NVIDIA GPU ⓘ |
| supportsDataType |
complex double-precision (for some routines)
ⓘ
complex single-precision (for some routines) ⓘ double-precision floating point ⓘ half-precision floating point (for some routines) ⓘ single-precision floating point ⓘ |
| supportsExecution | asynchronous GPU execution (via CUDA streams) ⓘ |
| supportsFeature |
batched sparse operations (for some routines)
ⓘ
descriptor-based matrix and vector objects ⓘ mixed-precision computation (for some routines) ⓘ |
| supportsLanguage |
C
ⓘ
C++ ⓘ Fortran (via wrappers) ⓘ |
| supportsMatrixFormat |
BSR (Block Sparse Row)
ⓘ
COO (Coordinate) ⓘ CSC (Compressed Sparse Column) ⓘ
linked to:
SparseMatrixCSC
CSR (Compressed Sparse Row) ⓘ ELL (ELLPACK) ⓘ HYB (Hybrid ELL/COO) ⓘ |
| supportsOperation |
sparse matrix factorization (limited routines)
ⓘ
sparse matrix format conversion ⓘ sparse matrix reordering ⓘ sparse matrix-matrix multiplication ⓘ sparse matrix-vector multiplication ⓘ sparse triangular solve ⓘ |
| targetPlatform |
NVIDIA GPU
ⓘ
linked to:
GPU
|
| useCase |
engineering simulation
ⓘ
graph analytics (via sparse matrices) ⓘ machine learning with sparse data ⓘ scientific computing ⓘ |
| vendor | NVIDIA Corporation ⓘ |
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: cuSPARSE Description of subject: cuSPARSE is NVIDIA’s GPU-accelerated library providing high-performance sparse linear algebra routines for CUDA applications.
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: cuSPARSE
subject linked to:
CUDA toolkit