Arm/inception-v3-int8-xnnpack-executorch-graviton-g4

Inception V3 optimized for Arm-based Cloud CPU

Inception V3 image classification quantized to INT8 with static post-training quantization (PTQ) and exported as an ExecuTorch .pte model for Arm-based Cloud CPU systems.

Summary

This repository contains an Arm-optimized version of TorchVision's inception_v3 for 1,000-class ImageNet classification. The model is provided in ExecuTorch .pte format (ExecuTorch runtime), targeting Cloud CPU systems.

This version demonstrates efficient inference on Arm-based platforms while preserving the original model's intended behavior. Arm evaluated this model on ImageNet-1k and measured performance on a representative evaluation target.

Key results

Area	Result
Model format	ExecuTorch .pte
Target device class	Cloud CPU
Reference device	AWS Graviton G4 (Neoverse-V2, Ubuntu 24.04.4 LTS)
Primary performance result	84.27 ms p50 latency (11.87 FPS), 1.32x faster than FP32
Accuracy result	Top-1 77.40% / Top-5 93.53%
Size / memory result	23.07 MB (.pte), 3.94x smaller than FP32

Original model

Field	Value
Original model	inception_v3
Original source	torchvision/models/inception.py
Original developer	Google for the architecture; PyTorch TorchVision for the pretrained model
Original model card	torchvision.models.inception_v3
Original license	BSD-3-Clause

Model files

File	Description
inception_v3_executorch_optimized.pte	Arm-optimized INT8 model for deployment; the file the example loads
example.py	Minimal inference example
pyproject.toml	Pinned runtime dependencies and deployment metadata
benchmarks/	FP32 baseline and Arm-optimized benchmark records

Performance

Performance was measured on the reference configuration below. Results are intended to make the optimization reproducible but do not guarantee identical performance on every Arm-based system.

Reference configuration

Field	Value
Device / platform	AWS Graviton G4
CPU / accelerator	Neoverse-V2, CPU
OS	Linux — Ubuntu 24.04.4 LTS
Runtime	ExecuTorch 1.1.0
Backend / delegate	XNNPACK + KleidiAI
Batch size	1
Input resolution	299x299
Precision	INT8 PTQ-static (symmetric per-channel weights, asymmetric per-tensor affine activations)
Runs	10 warmup + 100 measured

Performance results

Metric	Original / baseline	Arm-optimized	Improvement
p50 latency	111.22 ms	84.27 ms	1.32x
p90 latency	111.49 ms	84.35 ms	1.32x
p99 latency	112.31 ms	84.43 ms	1.33x
Frames per second	8.99 FPS	11.87 FPS	1.32x
Model load time	60.17 ms	22.69 ms	2.65x
Time to first inference	113.62 ms	86.58 ms	1.31x
Model size	90.97 MB	23.07 MB	3.94x smaller
Peak memory	116.96 MB	40.18 MB	2.91x less

Accuracy

Accuracy was evaluated using the same preprocessing, input resolution, and evaluation protocol described below. Where possible, the optimized model is compared against the original model under the same evaluation conditions.

Evaluation setup

Field	Value
Dataset	ImageNet-1k
Split	val
Number of samples	50000
Metric(s)	Top-1 Accuracy, Top-5 Accuracy
Evaluation runtime	ExecuTorch

Accuracy results

Metric	Original / baseline	Arm-optimized	Change
Top-1 Accuracy	77.51%	77.40%	-0.11 pp
Top-5 Accuracy	93.57%	93.53%	-0.04 pp

Accuracy was measured using the evaluation setup described above. Users should re-evaluate the model on their own data before production use.

Arm optimization approach

Arm optimized this model for efficient inference on Arm-based platforms using a hardware-aware conversion and validation flow.

For this release, Arm used:

Optimization area	Applied?	Notes
Model conversion	Yes	Converted to ExecuTorch .pte through the PT2E export path, with the auxiliary classifier disabled so the graph has a single output
Quantization	Yes	Static INT8 PTQ with symmetric per-channel weights and asymmetric per-tensor affine activations, calibrated on 500 randomly selected ImageNet-1k samples. No layers are excluded from quantization
Runtime/backend selection	Yes	XNNPACK + KleidiAI delegate
Graph/runtime compatibility updates	Yes	Performed as part of the ExecuTorch export pipeline
Accuracy validation	Yes	Compared against the original model under the same evaluation conditions
Performance validation	Yes	Measured on the reference Arm platform

The goal of this process is to improve deployment characteristics such as latency, memory use, model size, and runtime compatibility while preserving the model's intended behavior. Detailed conversion scripts, calibration configuration, or backend-specific implementation details may be provided separately where appropriate.

Using this model

Install dependencies

From this model directory:

uv python install
uv sync --frozen

Run the example

uv run example.py

The ordinary run prints its top-five predictions without changing repository files. To write a separate result and compare it with the committed expected result for sample_input.jpg:

uv run example.py --output /tmp/inception-v3-predictions.json
cmp /tmp/inception-v3-predictions.json predictions.json

Expected input

Property	Value
Input shape	[1, 3, 299, 299]
Input type	float32
Input range	[0.0, 1.0] before normalization
Preprocessing	Resize shorter edge to 342 (bilinear, antialiased), center-crop 299x299, ImageNet mean/std normalization

Expected output

Property	Value
Output shape	[1, 1000]
Output type	float32 (raw class logits)
Postprocessing	softmax, top-5

Intended use

This model is intended for developers evaluating image-classification workloads on Arm-based platforms. It is suitable as a reference implementation for benchmarking, prototyping, and integration exploration.

Limitations

• Performance depends on the target device, runtime version, backend/delegate support, memory configuration, and system load.
• Accuracy was evaluated on ImageNet-1k val and may not generalize to all domains.
• This release preserves the original model's intended task and behavior, but users should validate it for their own application, data, and deployment environment.
• This repository is not a replacement for the original model documentation.

Additional notes

• The exported model has a fixed 299x299 input resolution
• Sample input: sample_input.jpg is derived from Samoyed on Beach by Appleinfl, via Wikimedia Commons (public domain).

About this version

Original Model: inception_v3 by Google for the architecture; PyTorch TorchVision for the pretrained model - Repository

Optimization/conversion: Arm-Optimized version for execution on Arm-based platforms.

Converted/optimized by: Arm

License: The Original Model and the Optimized Model are subject to BSD-3-Clause.

This repository contains a converted or optimized version of the Original Model (the “Optimized Model”). The Original Model has been converted or optimized as described above for execution on Arm-based platforms.

No retraining or fine-tuning of the Original Model was performed as part of the conversion or optimization. The conversion or optimization was not intended to change the Original Model’s behavior or intended use.

Original Model and Documentation

For information about the Original Model, including its development, training data, intended uses, limitations and other relevant information, please refer to the Original Model repository. Information in that repository was provided by the original developer or other third parties and, unless expressly stated otherwise, has not been independently verified by Arm.

Licenses and Third-Party Terms

Use of the Original Model and the Optimized Model is subject to the applicable licenses, usage restrictions and other terms identified above and in the relevant repositories. Publication of the Optimized Model does not grant any rights beyond those provided under the applicable license terms.

You are responsible for reviewing those terms and ensuring that your use of the Original Model and the Optimized Model is permitted.

Purpose of this Release

The Optimized Model is provided as a reference implementation to demonstrate and evaluate execution and performance on Arm-based systems. It is not a production-ready or supported solution.

Arm’s publication of the Optimized Model does not constitute an endorsement or certification of the Original Model or a representation that the Optimized Model is suitable for production use or any particular purpose.

To the fullest extent permitted by applicable law (i) the Optimized Model is provided “as is.” Arm makes no representations or warranties that the Original Model, the Optimized Model or their outputs are accurate, safe, secure, non-infringing, legally compliant, suitable for production use or fit for any particular purpose; and (ii) Arm will not be liable for any loss or damage arising from or in connection with the Optimized Model, its use or its outputs.

You are responsible for independently evaluating the Optimized Model, its outputs and its suitability for your intended use, including compliance with applicable legal, regulatory, safety and security requirements.

Arm does not commit to provide ongoing support, maintenance or updates for the Optimized Model. Any use of or reliance on the Optimized Model or its outputs is at your own risk.