# Pytorch workflow

## Eager mode fine tune

The PyTorch stack supports eager mode execution for Cloud AI accelerators
via the `torch_qaic` package.

![../../../../../_images/pytorch-stack.png](data:image/png;base64,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)

- The `torch_qaic` Python package lists operators supported by Cloud
AI accelerators.
- `torch_qaic` registers the accelerator with PyTorch.
- The JIT runtime enables running operators on Cloud AI hardware.
- The operator library is pre-coded and supports multiple neural signal processors (NSPs).

The [Qualcomm Efficient-Transformers Library](https://quic.github.io/efficient-transformers/source/release_docs.html) provides the
necessary infrastructure for fine-tuning models. For more details, refer
to [Finetune
Infra](https://quic.github.io/efficient-transformers/source/finetune.html).

### Limitations

- x86 host support only
- Python 3.10 only
- Torch 2.4.1 only
- Precision support: FP32, FP16, Int32, Int64 (assuming data is within
Int32 range), Boolean

Last Published: May 01, 2026

[Previous Topic
Execute the QPC](https://docs.qualcomm.com/bundle/publicresource/80-99100-3/topics/index_model-execution.md) [Next Topic
Model architecture](https://docs.qualcomm.com/bundle/publicresource/80-99100-3/topics/index_Model-Architecture-Support.md)