Open-source profiling tool gives developers cycle-accurate insight to optimize AI performance and energy efficiency on Ambiq Apollo SoCs.
AUSTIN, TX— Ambiq Micro has announced the launch of heliaPROFILER™, a new open-source profiling tool that expands its HELIA™ AI ecosystem. Designed for AI developers building on Ambiq Apollo SoCs, heliaPROFILER provides cycle-accurate insight into AI model execution on production hardware, enabling faster optimization of performance, memory usage, and energy efficiency.
As AI models become increasingly sophisticated, understanding how they perform on resource-constrained edge devices has become essential. Yet profiling AI workloads on embedded hardware often requires fragmented workflows and manual analysis that slow development. heliaPROFILER streamlines that process into a single automated workflow designed to provide cycle-accurate visibility into model performance, memory usage, and optional real-time power measurement on Ambiq Apollo SoCs. This will help developers quickly identify performance bottlenecks, compare optimization strategies, and bring production-ready Edge AI applications to market faster.
“Ambiq’s ultra-low-power silicon is the foundation for deploying AI in always-on devices,” said Dr. Adam Page, Head of AI at Ambiq. “heliaPROFILER will give developers actionable insight into how AI workloads execute on Apollo hardware, helping them improve performance and take advantage of the silicon’s energy efficiency. By connecting AI optimization directly to production hardware, we are making it easier for customers to move from model development to deployment.”
heliaPROFILER is the latest addition to Ambiq’s growing HELIA AI software ecosystem. Together with heliaCORE™, heliaRT™, and heliaAOT™, it provides developers with a more complete software stack for building, optimizing, and deploying AI applications on ultra-low power edge devices.
Key Capabilities:
- Automated profiling from build through execution and reporting with a single command
- Layer-level performance analysis to quickly identify runtime bottlenecks
- Runtime comparison across heliaRT™ and heliaAOT™ execution engines
- Energy-aware optimization with optional real-time power measurement
- Visual performance reporting compatible with Google’s Model Explorer
Released as an open-source project, heliaPROFILER is available today as a Python package on PyPI (pip install helia-profiler) and on GitHub. The current alpha release supports Ambiq Apollo platforms, with additional platform support planned as the project advances toward version 1.0.
Learn more about heliaPROFILER here.
