---
tags: Concept, CCS, ZH
title: CCS Image-TensorRT | zh
GA: UA-155999456-1
---
{%hackmd @docsharedstyle/default %}
{%hackmd @docsharedstyle/twccheader-zh %}
{%hackmd @docsharedstyle/ccs-image-icon-size %}
# <img class="ccsimgicon" src="https://cos.twcc.ai/SYS-MANUAL/uploads/upload_f55059e9d0a6ac45c44bcc0ec1bebff5.png"> TensorRT
TWCC 提供 NGC’s TensorRT 立即可用的工作環境。NVIDIA TensorRT 是一個 C++ 函式庫，可以加快在 NVIDIA GPU 上的高效能推理運算，能結合常用於訓練的深度學習框架。TensorRT 致力於在 GPU 上快速有效地運行已訓練過的神經網路，並產生推理的結果。
## <i class="fa fa-sticky-note" aria-hidden="true"></i> <span class="ccsimglist">映像檔版本資訊</span>
:::info
{%hackmd @docsharedstyle/note-zh %}
`py3` 與 `py2` 則為 Python 之版本差異。
:::
<details class="docspoiler">
<summary><b>點我參考詳細版本套件資訊</b></summary>
- [tensorrt-19.08-py2/py3](https://docs.nvidia.com/deeplearning/tensorrt/container-release-notes/rel_19-08.html#rel_19-08)
- [tensorrt-19.02-py2/py3-v1](https://docs.nvidia.com/deeplearning/tensorrt/container-release-notes/rel_19-02.html#rel_19-02)
- [tensorrt-18.12-py2/py3-v1](https://docs.nvidia.com/deeplearning/tensorrt/container-release-notes/rel_18.12.html#rel_18.12)
- [tensorrt-18.10-py2/py3-v1](https://docs.nvidia.com/deeplearning/tensorrt/container-release-notes/rel_18.10.html#rel_18.10)
- [tensorrt-18.08-py2/py3-v1](https://docs.nvidia.com/deeplearning/tensorrt/container-release-notes/rel_18.08.html#rel_18.08)
</details>