Qualcomm®  AI Inference Suite

A comprehensive set of ready-to-use AI applications, agents, tools, and libraries for developing and deploying AI inference on premises or via cloud deployments.  

The Qualcomm^® AI Inference Suite comprises a Python SDK and OpenAI-compatible APIs. These interfaces simplify deployment of AI applications and agents powered by Qualcomm^® Cloud AI inference accelerator cards to achieve industry-leading performance per watt at a low total-cost-of-ownership (TCO).

Use pre-trained generative AI models for chat, generative multimedia, and retrieval-augmented generation (RAG).

Product license agreement

Featured ResourcesGetting Started Guide

REST API Reference Guide

Qualcomm Cloud AI family of inference accelerators

Tutorials

Get Software

1. Run:

pip install python-imagine-sdk

2. Select an AI inference playground and generate an API key.

3. Visit our Getting Started Guide.

Qualcomm Dev Playground Intro

02:13

2:13

Video Player is loading.

Current Time 0:00

Duration 2:12

Loaded: 4.48%

Stream Type LIVE

Remaining Time 2:12

Developer Playgrounds

The region-specific AI inference playgrounds below provide free-to-test examples, documentation, and other learning resources so you can immediately experiment with AI inference.

Cirrascale:
United States

powered by Qualcomm Cloud AI

Try it now

Core42:
United Arab Emirates

powered by Qualcomm Cloud AI

Playground access

ALLaM Developer Playground:
Saudi Arabia

powered by Qualcomm Cloud AI

Playground access

Cloud AI Inference Services

Ready to deploy your AI applications at scale? Qualcomm Cloud AI, as part of cloud operators’ inference services, delivers the performance and power efficiency necessary to deploy and accelerate AI inference at scale.

Try the Cirrascale Inference Cloud

On-Prem Appliances

Qualcomm Dragonwing™ AI On-Prem Appliance Solution with Qualcomm Cloud AI cards and Qualcomm AI Inference Suite is designed for generative AI inference and computer vision workloads on dedicated on-premises hardware. This setup allows your sensitive customer data, fine-tuned models, and inference loads to remain on premises.

Learn more about our on-prem appliances

The Qualcomm Cloud AI Inference Suite is built upon lower-level frameworks and runtimes, allowing you to bring your own model if needed.

View documentation

Playground GUI

Try out the inference cloud  in a web browser.

Python SDK

Easily use models in the  Qualcomm AI Inference Suite directly with Python.

Learn more

REST API

Use the REST API with  examples in your choice of programming language.

AI Inference with Ease

Gain immediate access to the playgrounds and choose from a wide selection of AI models, agents, application examples, and inference services.

Top Performance and Efficiency

Use software optimized for inference on Qualcomm Cloud AI inference accelerators for leading performance and lower total cost of ownership.

Flexible Deployments at Scale

Deploy in the cloud or to on-prem appliances supported by multi-tenancy, containerization, and auto-scaling with Kubernetes.

Robust Developer Options

Programmatically interact using the Python SDK, OpenAI-compatible APIs, and familiar AI frameworks like LangChain.

Application and Agent Samples

The Qualcomm AI Inference Suite supports a variety of models for common AI inference scenarios. Once you log in to one of the providers above, you’ll find samples and code you can use.

Agent examples using the Qualcomm AI Inference Suite are provided as starting samples to understand common scenarios. For example: a research agent that can pull together content from multiple locations, chatbots based on current events, or a RAG pattern for constantly updated source data.