# Cloud AI SDK

The Cloud AI SDK enables you to optimize trained deep learning models for high-performance inference on [Qualcomm AI platforms](https://docs.qualcomm.com/doc/80-99100-3/topic/index_Getting-Started.html#cloud-ai-platforms) with Qualcomm® Cloud AI 100 accelerators.

> 
> 
> 

## Get started

> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> Cloud AI SDK overview
> 
> 
> Get the essential details about the purpose of the Cloud AI SDK, its components, and platform support
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/index_Getting-Started.html
> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> Verify setup
> 
> 
> Verify the setup of the Qualcomm AI On-Prem Appliance plus installation instructions for your reference.
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/verify-setup.html
> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> Quick start guide
> 
> 
> Steps to run a sample model on Qualcomm Cloud AI Platforms.
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/index_Quick-Start-Guide.html

## User guide

> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> Inference workflow
> 
> 
> Explore the steps to prepare, export, compile, and deploy pre-trained models in production.
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/index_Inference-Workflow.html
> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> Pytorch workflow
> 
> 
> Execute PyTorch models in Eager Mode using `torch_qaic`.
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/index_Eager-Mode-Finetune.html
> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> Model architecture
> 
> 
> Cloud AI 100 inference cards support a wide range of model architectures.
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/index_Model-Architecture-Support.html
> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> Advanced SDK capabilities
> 
> 
> Specialized Cloud AI SDK functionality is available using [custom operations](https://docs.qualcomm.com/doc/80-99100-3/topic/index_custom_ops.html#custom-ops) and [model sharding](https://docs.qualcomm.com/doc/80-99100-3/topic/index_model_sharding.html#reference-to-model-sharding).
> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> System management
> 
> 
> Query card and SoC health using `qaic-util` and use AIC-manager to collect metrics.
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/index_System-Management.html
> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> Architecture
> 
> 
> Architecture details about the Cloud AI 100 inference cards.
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/index_Architecture.html

## APIs

> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> Python
> 
> 
> Python API reference for Qualcomm Cloud AI.
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/index_Python-API.html
> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> CPP API
> 
> 
> C++ API reference for Qualcomm Cloud AI.
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/index_Cpp-API.html
> 
> 
> <svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewbox="0 0 16 16" fill="none" aria-label="icon3">
>   <path d="M8 2V14M3.33333 2H12.6667C13.403 2 14 2.59695 14 3.33333V12.6667C14 13.403 13.403 14 12.6667 14H3.33333C2.59695 14 2 13.403 2 12.6667V3.33333C2 2.59695 2.59695 2 3.33333 2Z" stroke="#2A2AEA" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"></path>
> </svg> ONNX Runtime
> 
> 
> ONNX Runtime reference for Qualcomm Cloud AI.
> 
> https://docs.qualcomm.com/doc/80-99100-3/topic/index_onnxrt-qaic.html

Last Published: May 01, 2026

[Next Topic
Cloud AI SDK overview](https://docs.qualcomm.com/bundle/publicresource/80-99100-3/topics/index_Getting-Started.md)