# Fine tune model accuracy with Qualcomm AI Model Efficiency Toolkit (AIMET)

AIMET is a software toolkit for quantizing trained ML models. It improves the
runtime performance of deep learning models and facilitates their deployment
on edge devices, like mobile phones or laptops, by reducing the model’s compute
load and memory footprint, as shown in the following image.

![../_images/aimet.png](data:image/png;base64,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)

AIMET employs post-training and fine tuning techniques to minimize accuracy loss during
quantization and compression and supports models from the ONNX and PyTorch frameworks.

For details on using AIMET to improve model accuracy, see the
[AIMET documentation](https://quic.github.io/aimet-pages/releases/latest/index.html).

Last Published: May 14, 2026

[Previous Topic
Fine tune an AI model with custom data using Edge Impulse](https://docs.qualcomm.com/bundle/publicresource/80-80022-15B/topics/edge-impulse.md) [Next Topic
Use available frameworks and runtimes](https://docs.qualcomm.com/bundle/publicresource/80-80022-15B/topics/use-available-frameworks-and-runtimes.md)