# 设置 LiteRT 以优化 AI 模型

![../_images/optimize-compile-tflite.png](data:image/png;base64,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)

LiteRT 是一个用于设备推理的开源深度学习框架。LiteRT 通过优化模型的延迟、模型大小、功耗等，帮助开发者在移动、嵌入式和边缘平台上运行他们的模型。Qualcomm 支持通过以下列出的 LiteRT Delegate 在 Qualcomm Linux 硬件上本地执行 LiteRT 模型。

| **Delegate** | **加速** |
| --- | --- |
| AI Engine Direct Delegate (QNN Delegate) | CPU、GPU 和 HTP |
| XNNPack Delegate | CPU |
| GPU Delegate | GPU |

<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewbox="0 0 800 500" width="800" height="500" style="cursor:auto !important" aria-label="../_images/ai-hub-video.svg" svgdefaultwidth="75%">
    <defs>
      <style>@import url("https://fonts.googleapis.com/css2?family=Roboto+Flex:opsz,wght@8..144,100..1000&amp;display=swap");
.svg-1 .bg-fill { fill: var(--color-background) }
.svg-1 .fill-text { color: var(--color-content); fill: var(--color-content) }
.svg-1 .video-hoverbox { transition: opacity 0.15s ease-in-out }
.svg-1 .video-hoverbox:hover { opacity: 0.9 }</style>
  </defs>

  <foreignobject x="0" y="0" width="800" height="500">
    <body xmlns="http://www.w3.org/1999/xhtml">
        <iframe width="800" height="500" src="https://players.brightcove.net/1414329538001/4JiZQnWhg_default/index.html?videoId=6366848482112" allowfullscreen="" allow="encrypted-media"></iframe>
    <div class='topic-detail'><div class='topic-updated-date'><span> Last Published: </span>Oct 12, 2025</div><div class='prev-and-next-links'><span class='previous-topic-link'><span aria-hidden='true' class='disabled' data-tip='' data-effect='solid'></span></span></div></div></body>
    </foreignobject>
</svg>

参考指南

![arrow-right](data:image/png;base64,UklGRpwAAABXRUJQVlA4TJAAAAAvF8AFEH+goG0bxqU4/kwuDQNp22T7Hd2/tito24Zxx5/keDw1kaw6pAoFSMK/iB/qXPECxHNdR35a5LyfAAqFBeA/Hm9S6E2CUWxbbf5gABvsO2yRgABSJERZJERCJMRBb/sM1UBE/xWmbcM46e4lUEfjF2LpUYCVX6HiHS5YDP3Jhgk/hl+K1nlAASYCVQQ=) [参考指南](https://docs.qualcomm.com/bundle/publicresource/topics/80-70020-54/tflite-landing-page.html)

运行示例程序

![arrow-right](data:image/png;base64,UklGRpwAAABXRUJQVlA4TJAAAAAvF8AFEH+goG0bxqU4/kwuDQNp22T7Hd2/tito24Zxx5/keDw1kaw6pAoFSMK/iB/qXPECxHNdR35a5LyfAAqFBeA/Hm9S6E2CUWxbbf5gABvsO2yRgABSJERZJERCJMRBb/sM1UBE/xWmbcM46e4lUEfjF2LpUYCVX6HiHS5YDP3Jhgk/hl+K1nlAASYCVQQ=) [运行](https://docs.qualcomm.com/bundle/publicresource/topics/80-70020-50/ai-ml-sample-applications.html)

API 手册

![arrow-right-up](data:image/png;base64,UklGRnoAAABXRUJQVlA4TG4AAAAvF8AFEDdAIAiRO/7v4waBIETu+L8PAUni/6l9CBS0bcOUepmeDcYLoO0x6J8S2iYYtY0kKVIjmAvAQJg9IGzjaP40tqdeAyCi/xPASXsppcxs/EoGRFd8CdFN8MUuugm+2EU3wRdCN+UF+LJDAg==) [C/C++](https://www.tensorflow.org/lite/api_docs/cc?_gl=1%2Axotull%2A_ga%2AOTM5NzE4ODUwLjE2ODk2OTQ3MDI.%2A_ga_W0YLR4190T%2AMTcxNTM3NzQyMS4yNy4xLjE3MTUzNzc0MjEuMC4wLjA)

Last Published: Oct 12, 2025

[Previous Topic
使用 AI Hub 优化 AI 模型](https://docs.qualcomm.com/bundle/publicresource/80-70020-15BY/topics/ai-hub.md) [Next Topic
使用 Qualcomm AI Runtime SDK 优化 AI 模型](https://docs.qualcomm.com/bundle/publicresource/80-70020-15BY/topics/qairt.md)