Support Matrix#
These support matrices provide an overview of the supported platforms, features, and hardware capabilities of the TensorRT APIs, parsers, and layers.
Features for Platforms and Software#
Software |
Linux x86-64 |
Windows x64 |
Linux SBSA |
JetPack |
---|---|---|---|---|
10.9.x |
10.9.x |
10.9.x |
10.7.x |
|
NVIDIA cuDNN (Optional) |
||||
TensorRT Python API |
Yes |
Yes |
Yes |
Yes |
NvOnnxParser |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Note
Serialized engines are not portable across platforms.
If a serialized engine was created using the version-compatible flag, it could run with newer versions of TensorRT within the same major version.
If a serialized engine was created with hardware compatibility mode enabled, it can run on more than one kind of GPU architecture; the specifics depend on the hardware compatibility level used. Otherwise, serialized engines are not portable across devices.
Refer to the minimum compatible driver versions in the NVIDIA CUDA Release Notes for specific NVIDIA Driver versions.
Hardware and Precision#
The following table lists NVIDIA hardware and the precision modes each hardware supports. TensorRT has been compiled to support all NVIDIA hardware with SM 7.5 or higher capability. The table also lists the availability of DLA on this hardware. Refer to the following tables for the specifics.
Example Devices |
TF32 |
FP32 |
FP16 |
FP8 |
FP4 |
BF16 |
INT8 |
FP16 Tensor Cores |
INT8 Tensor Cores |
DLA |
|
---|---|---|---|---|---|---|---|---|---|---|---|
12.0 |
NVIDIA RTX 5090 |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
No |
10.0 |
NVIDIA B200 |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
No |
9.0 |
NVIDIA H100 |
Yes |
Yes |
Yes |
Yes |
Yes [5] |
Yes |
Yes |
Yes |
Yes |
No |
9.0 |
NVIDIA GH200 480 GB |
Yes |
Yes |
Yes |
Yes |
Yes [5] |
Yes |
Yes |
Yes |
Yes |
No |
8.9 |
NVIDIA L40S |
Yes |
Yes |
Yes |
Yes |
Yes [5] |
Yes |
Yes |
Yes |
Yes |
No |
8.7 |
NVIDIA DRIVE AGX Orin |
Yes |
Yes |
Yes |
No |
No |
No |
Yes |
Yes |
Yes |
Yes |
8.6 |
NVIDIA A10 |
Yes |
Yes |
Yes |
No |
No |
Yes |
Yes |
Yes |
Yes |
No |
8.0 |
NVIDIA A100 |
Yes |
Yes |
Yes |
No |
No |
Yes |
Yes |
Yes |
Yes |
No |
7.5 |
NVIDIA T4 |
No |
Yes |
Yes |
No |
No |
No |
Yes |
Yes |
Yes |
No |
Compute Capability Per Platform#
The section lists the supported computing capability based on the platform.
Software Versions Per Platform#
Platform |
Compiler Version |
Python Version |
---|---|---|
Ubuntu 20.04 x86-64 |
||
Ubuntu 22.04 x86-64 |
||
Ubuntu 24.04 x86-64 |
||
Rocky Linux 8.9 x86-64 |
||
Rocky Linux 9.3 x86-64 |
||
SLES 15 x86-64 |
N/A |
|
|
N/A |
|
Ubuntu 24.04 SBSA |
||
NVIDIA JetPack AArch64 |
Note
The Debian or RPM packages support the Python version listed in the table. Python versions 3.8, 3.9, 3.10, 3.11, 3.12, and 3.13 are supported using Python wheel files from the tar or zip packages.
ONNX Operator Support#
The ONNX operator support list for TensorRT can be found here.
Footnotes