Hexagon NN library
E660969
Hexagon NN library is a neural network software framework optimized for Qualcomm's Hexagon DSP architecture to accelerate on-device AI inference.
All labels observed (1)
| Label | Occurrences |
|---|---|
| Hexagon NN library canonical | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T7388446 — 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: Hexagon NN library Context triple: [Hexagon DSP, hasSoftwareTool, Hexagon NN library]
-
A.
NVIDIA TensorRT
NVIDIA TensorRT is a high-performance deep learning inference optimizer and runtime library designed to accelerate AI models on NVIDIA GPUs in production environments.
-
B.
NVIDIA Triton Inference Server
NVIDIA Triton Inference Server is an open-source, production-ready platform for serving and scaling AI model inference across GPUs and CPUs with support for multiple frameworks and deployment environments.
-
C.
ONNX
ONNX (Open Neural Network Exchange) is an open standard format for representing machine learning models that enables interoperability between different deep learning frameworks and tools.
-
D.
PlaidML
PlaidML is an open-source, hardware-agnostic deep learning engine designed to accelerate neural network computation on a wide range of GPUs and other devices.
-
E.
ShuffleNetV2
ShuffleNetV2 is a lightweight convolutional neural network architecture designed for efficient image classification on resource-constrained devices, emphasizing speed and low computational cost.
- 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: Hexagon NN library Target entity description: Hexagon NN library is a neural network software framework optimized for Qualcomm's Hexagon DSP architecture to accelerate on-device AI inference.
-
A.
NVIDIA TensorRT
NVIDIA TensorRT is a high-performance deep learning inference optimizer and runtime library designed to accelerate AI models on NVIDIA GPUs in production environments.
-
B.
NVIDIA Triton Inference Server
NVIDIA Triton Inference Server is an open-source, production-ready platform for serving and scaling AI model inference across GPUs and CPUs with support for multiple frameworks and deployment environments.
-
C.
ONNX
ONNX (Open Neural Network Exchange) is an open standard format for representing machine learning models that enables interoperability between different deep learning frameworks and tools.
-
D.
PlaidML
PlaidML is an open-source, hardware-agnostic deep learning engine designed to accelerate neural network computation on a wide range of GPUs and other devices.
-
E.
ShuffleNetV2
ShuffleNetV2 is a lightweight convolutional neural network architecture designed for efficient image classification on resource-constrained devices, emphasizing speed and low computational cost.
- F. None of above. chosen
Statements (49)
| Predicate | Object |
|---|---|
| instanceOf |
AI inference runtime
ⓘ
neural network software framework ⓘ |
| accessMethod | Qualcomm developer tools and SDKs ⓘ |
| designedFor |
edge devices
ⓘ
mobile devices ⓘ |
| developer | Qualcomm NERFINISHED ⓘ |
| documentation | Qualcomm developer network NERFINISHED ⓘ |
| feature |
hardware-aware kernel optimizations
ⓘ
low-latency inference ⓘ power-efficient execution ⓘ support for parallel execution on DSP cores ⓘ |
| integratesWith |
Qualcomm AI Engine
NERFINISHED
ⓘ
Qualcomm Snapdragon mobile platforms NERFINISHED ⓘ |
| license | proprietary software ⓘ |
| optimizedFor |
Qualcomm Hexagon DSP architecture
NERFINISHED
ⓘ
on-device AI inference ⓘ |
| partOf | Qualcomm AI software stack NERFINISHED ⓘ |
| provides |
APIs for deploying neural network models
ⓘ
graph-based neural network execution ⓘ tools for compiling models to Hexagon DSP ⓘ |
| purpose |
accelerate AI workloads on-device
ⓘ
improve power efficiency for AI inference ⓘ offload AI computation from CPU ⓘ reduce latency of neural network inference ⓘ |
| relatedTo |
Hexagon Vector eXtensions (HVX)
NERFINISHED
ⓘ
Qualcomm Neural Processing SDK NERFINISHED ⓘ |
| runsOn |
Hexagon DSP
NERFINISHED
ⓘ
Qualcomm Snapdragon SoCs with Hexagon DSP ⓘ |
| supports |
convolutional neural networks
ⓘ
deep neural network inference ⓘ neural network layers such as activation functions ⓘ neural network layers such as convolution ⓘ neural network layers such as fully connected layers ⓘ neural network layers such as pooling ⓘ quantized neural networks ⓘ |
| supportsDataType |
16-bit fixed-point arithmetic
ⓘ
32-bit floating point (in some configurations) ⓘ 8-bit integer quantization ⓘ |
| supportsModelImportFrom |
Caffe via conversion tools
ⓘ
TensorFlow via conversion tools ⓘ |
| supportsOptimization |
layer fusion
ⓘ
memory bandwidth reduction ⓘ vectorized DSP instructions ⓘ |
| targetUseCase |
augmented reality applications
ⓘ
on-device computer vision ⓘ on-device natural language processing ⓘ on-device speech recognition ⓘ |
| usedIn |
Android smartphones with Snapdragon processors
ⓘ
IoT devices using Snapdragon platforms ⓘ |
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: Hexagon NN library Description of subject: Hexagon NN library is a neural network software framework optimized for Qualcomm's Hexagon DSP architecture to accelerate on-device AI inference.
Referenced by (1)
Full triples — surface form annotated when it differs from this entity's canonical label.