Third-party support matrix

Applies to Linux

2024-03-22

3 min read time

ROCm™ supports various third-party libraries and frameworks. We’ve tested our supported versions, so you can be assured that they work. Non-supported third-party versions may also work, but we haven’t tested these for functionality.

Deep learning

ROCm releases support the most recent and two prior releases of PyTorch and TensorFlow. For TensorFlow the exact version follows the <TensorFlowVersion>.<ROCmVersionWithoutCommas> format.

ROCm	Pytorch	TensorFlow
5.0.2	1.8, 1.9, 1.10	2.6, 2.7, 2.8
5.1.3	1.9, 1.10, 1.11	2.7, 2.8, 2.9
5.2.x	1.10, 1.11, 1.12	2.8, 2.9, 2.9
5.3.x	1.10.1, 1.11, 1.12.1, 1.13	2.8, 2.9, 2.10
5.4.x	1.10.1, 1.11, 1.12.1, 1.13	2.8, 2.9, 2.10, 2.11
5.5.x	1.11, 1.12.1, 1.13, 2.0	2.10, 2.11, 2.12
5.6.x	1.12.1, 1.13, 2.0, 2.1	2.12, 2.13, 2.14
5.7.x	1.12.1, 1.13, 2.0, 2.1	2.12, 2.13, 2.14
6.0.x	1.13, 2.0, 2.1	2.12, 2.13, 2.14

Communication libraries

ROCm supports OpenUCX, an open-source, production-grade communication framework for data-centric and high performance applications.

UCX version	ROCm 5.4 and older	ROCm 5.5 and newer
-1.14.0	COMPATIBLE	INCOMPATIBLE
1.14.1+	COMPATIBLE	COMPATIBLE

The Unified Collective Communication (UCC) library also has support for ROCm devices.

UCC version	ROCm 5.5 and older	ROCm 5.6 and newer
-1.1.0	COMPATIBLE	INCOMPATIBLE
1.2.0+	COMPATIBLE	COMPATIBLE

Algorithm libraries

ROCm releases provide algorithm libraries with interfaces compatible with contemporary CUDA/NVIDIA HPC SDK alternatives.

• Thrust → rocThrust

• CUB → hipCUB

ROCm	Thrust / CUB	HPC SDK
5.0.2	1.14	21.9
5.1.3	1.15	22.1
5.2.x	1.15	22.2, 22.3
5.3.x	1.16	22.7
5.4.x	1.16	22.9
5.5.x	1.17	22.9
5.6.x	1.17.2	22.9
5.7.x	1.17.2	22.9
6.0.x	2.0.1	22.9

For the latest documentation of these libraries, refer to the ROCm API libraries.