onnxruntime-ep-nv-tensorrt-rtx-cu13 0.3.0__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
|
@@ -0,0 +1,94 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: onnxruntime-ep-nv-tensorrt-rtx-cu13
|
|
3
|
+
Version: 0.3.0
|
|
4
|
+
Summary: ONNX Runtime plugin package for the NVIDIA TensorRT RTX execution provider (EP ABI)
|
|
5
|
+
Author: NVIDIA Corporation
|
|
6
|
+
License-Expression: Apache-2.0
|
|
7
|
+
Project-URL: Homepage, https://github.com/NVIDIA/TensorRT-RTX-EP-ABI
|
|
8
|
+
Keywords: onnxruntime,tensorrt,execution-provider,nvidia
|
|
9
|
+
Classifier: Development Status :: 3 - Alpha
|
|
10
|
+
Classifier: Intended Audience :: Developers
|
|
11
|
+
Classifier: Programming Language :: Python :: 3
|
|
12
|
+
Classifier: Programming Language :: Python :: 3.8
|
|
13
|
+
Classifier: Programming Language :: Python :: 3.9
|
|
14
|
+
Classifier: Programming Language :: Python :: 3.10
|
|
15
|
+
Classifier: Programming Language :: Python :: 3.11
|
|
16
|
+
Classifier: Programming Language :: Python :: 3.12
|
|
17
|
+
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
|
|
18
|
+
Requires-Python: >=3.8
|
|
19
|
+
Description-Content-Type: text/markdown
|
|
20
|
+
Provides-Extra: dev
|
|
21
|
+
Requires-Dist: onnx>=1.14; extra == "dev"
|
|
22
|
+
Requires-Dist: numpy>=1.21; extra == "dev"
|
|
23
|
+
Dynamic: description
|
|
24
|
+
Dynamic: description-content-type
|
|
25
|
+
|
|
26
|
+
# onnxruntime-ep-nv-tensorrt-rtx
|
|
27
|
+
|
|
28
|
+
NVIDIA TensorRT RTX Execution Provider plugin for [ONNX Runtime](https://onnxruntime.ai/).
|
|
29
|
+
|
|
30
|
+
Enables hardware-accelerated inference on NVIDIA RTX GPUs (Ampere / RTX 30xx and later) via the
|
|
31
|
+
[ORT Plugin EP ABI](https://onnxruntime.ai/docs/execution-providers/plugin-ep-libraries.html).
|
|
32
|
+
|
|
33
|
+
---
|
|
34
|
+
|
|
35
|
+
## About NVIDIA TensorRT for RTX
|
|
36
|
+
|
|
37
|
+
NVIDIA® TensorRT™ for RTX (TensorRT-RTX) is an inference optimization library dedicated for
|
|
38
|
+
deploying AI inference on NVIDIA GeForce RTX GPUs. It is a great choice for developers building
|
|
39
|
+
applications that must run on Windows or Linux PCs, laptops, or workstations.
|
|
40
|
+
|
|
41
|
+
This package bundles the TensorRT-RTX runtime libraries alongside the ONNX Runtime EP plugin so
|
|
42
|
+
that no separate TensorRT-RTX installation is required.
|
|
43
|
+
|
|
44
|
+
For more information about TensorRT-RTX, visit https://developer.nvidia.com/tensorrt-rtx.
|
|
45
|
+
Online documentation: https://docs.nvidia.com/deeplearning/tensorrt-rtx/latest/index.html
|
|
46
|
+
License agreement: https://docs.nvidia.com/deeplearning/tensorrt-rtx/latest/reference/sla.html
|
|
47
|
+
|
|
48
|
+
---
|
|
49
|
+
|
|
50
|
+
## References
|
|
51
|
+
|
|
52
|
+
- Release Notes: https://docs.nvidia.com/deeplearning/tensorrt-rtx/latest/getting-started/release-notes.html
|
|
53
|
+
- Support Matrix: https://docs.nvidia.com/deeplearning/tensorrt-rtx/latest/getting-started/support-matrix.html
|
|
54
|
+
- Installation Guide: https://docs.nvidia.com/deeplearning/tensorrt-rtx/latest/installing-tensorrt-rtx/installation-overview.html
|
|
55
|
+
- C++ API: https://docs.nvidia.com/deeplearning/tensorrt-rtx/latest/_static/cpp-api/index.html
|
|
56
|
+
- Python API: https://docs.nvidia.com/deeplearning/tensorrt-rtx/latest/_static/python-api/index.html
|
|
57
|
+
|
|
58
|
+
---
|
|
59
|
+
|
|
60
|
+
## Requirements
|
|
61
|
+
|
|
62
|
+
- NVIDIA RTX GPU (Ampere or later)
|
|
63
|
+
- NVIDIA GPU driver with CUDA 13 support
|
|
64
|
+
- `pip install onnxruntime>=1.24`
|
|
65
|
+
|
|
66
|
+
## Installation
|
|
67
|
+
|
|
68
|
+
```bash
|
|
69
|
+
pip install onnxruntime>=1.24
|
|
70
|
+
pip install onnxruntime-ep-nv-tensorrt-rtx
|
|
71
|
+
```
|
|
72
|
+
|
|
73
|
+
## Usage
|
|
74
|
+
|
|
75
|
+
```python
|
|
76
|
+
import onnxruntime as ort
|
|
77
|
+
import onnxruntime_ep_nv_tensorrt_rtx as trt_ep
|
|
78
|
+
|
|
79
|
+
# Register the EP plugin
|
|
80
|
+
ort.register_execution_provider_library(trt_ep.get_ep_name(), trt_ep.get_library_path())
|
|
81
|
+
|
|
82
|
+
# List available devices
|
|
83
|
+
devices = [d for d in ort.get_ep_devices() if d.ep_name == trt_ep.get_ep_name()]
|
|
84
|
+
print(f"TensorRT RTX devices: {len(devices)}")
|
|
85
|
+
|
|
86
|
+
# Create session with EP
|
|
87
|
+
so = ort.SessionOptions()
|
|
88
|
+
so.add_provider_for_devices(devices, {})
|
|
89
|
+
sess = ort.InferenceSession("model.onnx", sess_options=so)
|
|
90
|
+
```
|
|
91
|
+
|
|
92
|
+
## License
|
|
93
|
+
|
|
94
|
+
Apache 2.0. See [LICENSE](https://www.apache.org/licenses/LICENSE-2.0).
|
|
@@ -0,0 +1,10 @@
|
|
|
1
|
+
# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
|
|
2
|
+
# SPDX-License-Identifier: Apache-2
|
|
3
|
+
|
|
4
|
+
[build-system]
|
|
5
|
+
requires = ["wheel-stub"]
|
|
6
|
+
build-backend = "wheel_stub.buildapi"
|
|
7
|
+
|
|
8
|
+
[tool.wheel_stub]
|
|
9
|
+
index_url = "https://pypi.nvidia.com/"
|
|
10
|
+
include_cuda_debuginfo = true
|