Build AI applications.
Compose your firmware.
The AI SDK
Compose the capabilities your application needs.
Reusable modules connect inference, data capture, connectivity and the system services around them.
NSX resolves the modules you select and their required dependencies, then integrates them into your CMake project. HAL, BSP and device support provide the hardware foundation. Start with a bare-metal application, or add the FreeRTOS module when you need tasks.
Start here
From setup to running firmware.
Select a step to explore it. Click the active step to pause or resume.
Install the NSX CLI after setting up the host tools and compiler.
nsx doctor names every host tool it needs and what a missing one would cost you.
A complete app directory for a real board: nsx.yml, CMakeLists.txt, sources and the starter modules.
Modules resolve and lock, CMake generates, Ninja builds. The result is an ordinary project you own.
nsx flash builds if needed, then programs the board over SWD through the SEGGER tools nsx doctor checked.
nsx view opens the SWO viewer and streams the firmware's output back to your terminal.
Abbreviated examples: install with uv tool install neuralspotx, check with nsx doctor,
create with nsx create-app, then nsx build, nsx flash and nsx view.
Hardware output is illustrative. Follow Getting started for the complete instructions.
Modules
Find a capability. Add it to your app.
A familiar dependency workflow, applied to embedded firmware.
Once your first app builds, add capabilities through NSX modules. Like pip, Cargo or npm, NSX uses named dependencies, manifests and a lockfile, with compatibility checks for your target.
nsx module search audionsx module add nsx-audionsx buildYour project
Select modules. Keep control of the build.
The generated project keeps your sources, board support and dependencies together.
After checking your environment, follow build, flash and view to run the app. App layout explains which files you own and which NSX regenerates.
HELIA
Add the AI capabilities your app needs.
Ambiq's AI stack reaches your app the same way every other capability does: as a module you add and lock.
Runtime
heliaRT
LiteRT for Micro / TFLM-compatible runtime with Reference, CMSIS-NN, and heliaCORE backends.
Kernels
heliaCORE
Optimized neural-network kernel library for Ambiq SoCs.
Compiler
heliaAOT
Ahead-of-time compiler producing standalone C inference modules.
heliaRT and heliaCORE reach your app as the nsx-helia-rt and nsx-cmsis-nn modules; both
also ship CMake and Zephyr integration in their public repositories, so choosing NSX here
is not the only way to consume them.
Read more: the KWS Inference example declares the
nsx-helia-rt module and builds like any other app here.
Explore the SDK
Find the modules and board support for your app.
The catalog is generated from registry and manifest data. Compatibility entries describe declared support.
Registry modules
50
Board descriptors
17
SoC families
11
Toolchains
3
Example apps
10
The packaged registry lists 50 modules. NSX ships 17 board descriptors across 11 SoC families, declaring 3 toolchains between them, and this repository carries 10 example apps.
Use the module catalog and board matrix to inspect the declarations, then read Boards and targets to configure your app. Check the module and example documentation for validation on your target.
In use
Start from a working example.
Explore complete NSX applications, with setup instructions and target support documented for each example.
Documentation
Find your way around.
Get started with NSX, explore development guides and modules, or look up commands and APIs.
Install NSX on macOS, Linux or Windows, check your environment and build your first board app.
Apps, modules, memory placement, startup and linker behavior, toolchains, and the concepts behind app generation.
The module registry: what each module provides, which SoCs and boards it declares compatibility with, and how to add it.
Every CLI command and option, the public Python API, and the schema of each manifest NSX reads.
Extend the SDK with custom modules or contribute a module. For automation, use the agent guidance, Markdown documentation and module catalog JSON.