# 0.5.0

## Release Information

- OS & Framework support - Tested with the following configurations:

> 
> 
> - Non-Gen AI:
> 
> 
> 
> > 
> > 
> > - Frameworks: ONNX, TFLite, Pytorch
> >         - Host Platforms: Linux-x86\_64 (Ubuntu 22.04), Windows-x86\_64 (10+), Windows-arm64 (10+)
> >         - Target Platforms: Linux-x86\_64 (Ubuntu 22.04), Linux-aarch\_64 (Ubuntu 22.04), Windows-arm64 (10+), Android-arm64, QNX
>     - Gen AI:
> 
> 
> 
> > 
> > 
> > - Frameworks: ONNX, GGUF
> >         - Host Platforms: Linux-x86\_64 (Ubuntu 22.04)
> >         - Target Platforms: Windows-arm64 (10+), Android-arm64
- QAIRT SDK - Tested with version (2.43.0, 2.44.0(Default))

Note

For QAIRT SDK, other supported SDK versions can be listed using `qairt-vm fetch --list`.

## Highlights

- Default QAIRT SDK updated to 2.44.0
- Using 2.44.0 QAIRT SDK adds the following features with QAIRT Dev Python APIs

> 
> 
> - AIMET Encodings V2 support for embedding lookup tables
>     - MHA2SHA reshape optimizations for Unet
>     - Custom IO creation via qairt.convert API
>     - 2-bit encoding support

## Known Issues

> 
> 
> - Access violation in `qairt.convert` on Windows on Snapdragon (WoS). Running `qairt.convert` results in a silent process exit due to an access violation. This is fixed in 0.6.0.
> - Models with batch size greater than 1
> 
> 
> 
> > 
> > 
> > - A discrepancy exists between the qnn-net-run CLI and the Python API regarding data preparation for batch sizes
> > greater than 1. While the CLI automatically groups individual tensor paths from an input list into a batch,
> > the Python API requires inputs to be pre-batched (concatenated into a single raw file) prior to execution
> - Gen AI Builder LoRA Workflow Failure: LoRA workflows may fail with a KeyError caused by an incorrect field in the lora\_importer\_config.yaml file.
> 
> 
> 
> > 
> > 
> > Workaround
> > 
> >     - Use a cache directory during generation.
> >     - After the failure occurs, manually edit lora\_importer\_config.yaml in the cache directory to replace “weights” with “lora\_weights”.
> >     - Resume Gen AI Builder with the updated configuration.

Last Published: May 26, 2026

[Previous Topic
Known Issues](https://docs.qualcomm.com/bundle/publicresource/80-87189-2/topics/0_4_0.md) [Next Topic
0.6.0](https://docs.qualcomm.com/bundle/publicresource/80-87189-2/topics/0_6_0.md)