# QNN CPU Op Support Revision History

| Introduced in QNN SDK Version | Description |
| --- | --- |
| 1.8.0 | <ul class="simple"><br><li><p>MatMul: Added support for default value for broadcast_dims parameter</p></li><br><li><p>Transform: Added support for QNN_DATATYPE_INT_32 data type</p></li><br><li><p>Cast: Added support for QNN_DATATYPE_UINT_8 and QNN_DATATYPE_INT_32 data types</p></li><br><li><dl class="simple"><br><dt>StridedSlice: Added support for QNN_DATATYPE_INT_32, QNN_DATATYPE_UINT_32,</dt><dd><p>QNN_DATATYPE_INT_8 and QNN_DATATYPE_UINT_8 data types</p><br></dd><br></dl><br></li><br><li><p>OneHot: Added support for QNN_DATATYPE_INT_32 data type for Input</p></li><br><li><p>Support added for ElementWiseCeil Op</p></li><br><li><p>Concat: Added support for QNN_DATATYPE_UINT_32 data type</p></li><br><li><p>ElementWiseMultiply: Added support for QNN_DATATYPE_UINT_32 data type</p></li><br><li><p>Support added for OneHot Op</p></li><br><li><p>AvgPool: Added support for count_pad_for_edges parameter</p></li><br><li><p>Support added for Unpack Op</p></li><br><li><p>ArgMin: Added support for keep_dim parameter</p></li><br><li><p>Support added for FloorDiv Op</p></li><br></ul> |

Last Published: Oct 10, 2025

[Previous Topic
Backend Specific Revision History](https://docs.qualcomm.com/bundle/publicresource/80-63442-50/topics/opdef_version_history.md) [Next Topic
QNN DSP Op Definition Revision History](https://docs.qualcomm.com/bundle/publicresource/80-63442-50/topics/dsp_opdef_version_history.md)