AIPC: Agent-Based Automation for AI Model Deployment with Qualcomm AI Runtime

TL;DR

AIPC is an agent-based system that automates AI model deployment on hardware, reducing manual effort and expertise needed, demonstrated with Qualcomm AI Runtime for various models.

Contribution

Introduces AIPC, an agent-driven framework that decomposes deployment into stages, injecting domain knowledge to automate and verify AI model deployment.

Findings

AIPC can deploy models from PyTorch to Qualcomm AI Runtime within 7-20 minutes.

Deployment costs range from approximately USD 0.7 to 10.

AIPC supports execution, failure localization, and repair for complex models.

Abstract

Edge AI model deployment is a multi-stage engineering process involving model conversion, operator compatibility handling, quantization calibration, runtime integration, and accuracy validation. In practice, this workflow is long, failure-prone, and heavily dependent on deployment expertise, particularly when targeting hardware-specific inference runtimes. This technical report presents AIPC (AI Porting Conversion), an AI agent-driven approach for constrained automation of AI model deployment. AIPC decomposes deployment into standardized, verifiable stages and injects deployment-domain knowledge into agent execution through Agent Skills, helper scripts, and a stage-wise validation loop. This design reduces both the expertise barrier and the engineering time required for hardware deployment. Using Qualcomm AI Runtime (QAIRT) as the primary scenario, this report examines automated deployment across representative vision, multimodal, and speech models. In the cases covered here, AIPC can complete deployment from PyTorch to runnable QNN/SNPE inference within 7-20 minutes for structurally regular vision models, with indicative API costs roughly in the range of USD 0.7-10. For more complex models involving less-supported operators, dynamic shapes, or autoregressive decoding structures, fully automated deployment may still require further advances, but AIPC already provides practical support for execution, failure localization, and bounded repair.