Benchmarks

OpenNN Performance Benchmark

Compare OpenNN against other machine learning frameworks across training speed, inference, memory, deployment size, dependencies, and energy consumption.

Build

No dependency stack and a low-memory runtime.

32x

Baseline RAM and GPU-ready VRAM

OpenNN vs PyTorch & TensorFlow · baseline footprint

Learn more

Zero

Dependencies & install friction

OpenNN vs PyTorch & TensorFlow · install requirements

Learn more

Learn

A compact, auditable C++ codebase and a concise API.

24x

Source lines of code

OpenNN vs PyTorch & TensorFlow · native source

Learn more

40%

Iris API lines of code

OpenNN vs PyTorch & TensorFlow · same Iris model

Learn more

Load

Memory-mapped data handling for larger datasets.

2.7x

Data capacity

OpenNN vs pandas + PyTorch/TensorFlow · same RAM

Learn more

Train

Faster training across dense, CNN, ResNet and transformer models.

1.79x

GPU HIGGS dense training

OpenNN vs PyTorch & TensorFlow · HIGGS

Learn more

1.20x

CPU HIGGS dense training

OpenNN vs PyTorch & TensorFlow · 3-run median, 0.7% dispersion

Learn more

1.69x

GPU Transformer training

OpenNN vs PyTorch & TensorFlow · GPU training

Learn more

1.7x

GPU ResNet-50 training

OpenNN vs PyTorch & TensorFlow · CIFAR-10

Learn more

Native

GPU on Windows

OpenNN vs PyTorch & TensorFlow · native CUDA path

Learn more

Optimize

Precision-aware acceleration across GPU training and inference.

2.37x

GPU fp32 vs bf16 precision sweep

OpenNN bf16 vs fp32 · best speedup across four workloads

Learn more

Validate

Matching accuracy, with better generalization.

On par

Numerical accuracy

OpenNN vs PyTorch & TensorFlow · held-out quality

Learn more

Deploy

Tiny self-contained runtimes, fast startup and dependency-free export.

1.5x

Deployment size on GPU (CNN)

OpenNN vs PyTorch & TensorFlow · CNN CUDA build

Learn more

138x

Deployment size on CPU

OpenNN vs PyTorch & TensorFlow · CPU package

Learn more

7x

Startup latency

OpenNN vs PyTorch & TensorFlow · first prediction

Learn more

Runtime-free

Model export to standalone code

OpenNN vs PyTorch & TensorFlow · standalone artifact

Learn more

Operate

Efficient inference with lower energy use in production.

Up to 1.52x

GPU Transformer inference

OpenNN vs PyTorch & TensorFlow · up to 680.4k tok/s

Learn more

Up to 1.42x

GPU HIGGS dense inference

OpenNN vs PyTorch & TensorFlow · 15.34M/34.84M samples/s

Learn more

1.21x

CPU HIGGS dense inference

OpenNN vs PyTorch & TensorFlow · 441.5k samples/s

Learn more

1.23-1.42x

GPU ResNet-50 inference

OpenNN vs PyTorch & TensorFlow · fp32/bf16, 5 runs

Learn more