# Build and run the Arm Machine Learning Evaluation Kit examples

## In this learning path

- [Introduction](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/mlek/)
- [Build the ML Evaluation Kit examples](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/mlek/build/)
- [Install Arm Ecosystem FVP](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/mlek/fvp/)
- [Run the examples on the FVP](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/mlek/run/)
- [Next Steps](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/mlek/_next-steps/)

## About this Learning Path

| Skill level:          | Introductory    |
|-----------------------|-----------------|
| Reading time:         | 30 min          |
| Last updated:         | 13 Aug 2026     |

| Author:               | Ronan Synnott, Arm [LinkedIn](https://linkedin.com/in/ronansynnott) |
|-----------------------|----------------------------------------------------------|
| Arm IP:               | [Cortex-M](https://support.arm.com/?tab=compute-ip&Product%20Type=Microcontrollers) [Ethos-U](https://support.arm.com/?tab=compute-ip&Product%20Type=Neural%20Processing%20Units) [Corstone](https://support.arm.com/?tab=compute-ip&Product%20Type=Subsystem%20Products) |
| Tags:                 | [ML](https://learn.arm.com/tag/ml) [Baremetal](https://learn.arm.com/tag/baremetal) [Arm Virtual Hardware](https://learn.arm.com/tag/arm-virtual-hardware) [FVP](https://learn.arm.com/tag/fvp) [GCC](https://learn.arm.com/tag/gcc) [Arm Compiler for Embedded](https://learn.arm.com/tag/arm-compiler-for-embedded) |

### Who is this for?

This is an introductory topic for embedded software developers interested in machine learning applications.

### What will you learn?

Upon completion of this Learning Path, you will be able to:

- Build examples from the Arm Machine Learning Evaluation Kit (MLEK)
- Run the examples on Arm Ecosystem Fixed Virtual Platform (FVP)

### Prerequisites

Before starting, you will need the following:

- Some familiarity with embedded programming
- A Linux host machine running Ubuntu

### Summary

You’ll build machine learning examples from MLEK and run them on the Corstone-320 FVP. You’ll locate generated AXF images, prepare the FVP, and launch an image with `-a`. You’ll configure the Ethos-U MAC count to match the build and verify that the application starts successfully.

### Frequently asked questions

<details>
<summary>How do I know the build succeeded, and where are the output binaries?</summary>
The build produces `.axf` images in a `cmake-*/bin` directory that reflects your configuration. List the files with `ls *.axf` to confirm the outputs exist.
</details>

<details>
<summary>Which FVP should I install to run these examples?</summary>
Install the Corstone-320 Ecosystem FVP on your local machine. Download it from the Arm Developer website and follow the referenced [Fast Model and FVP install guide](https://learn.arm.com/install-guides/fm_fvp/).
</details>

<details>
<summary>How can I shorten a long FVP command line?</summary>
Put the model configuration options in a file such as `config.txt`. Remove `-C` from each option in the file, then pass the file to the FVP with `-f config.txt`.
</details>

<details>
<summary>How do I load a specific example into the FVP?</summary>
Use the `-a` option to pass your desired `.axf` image location. Select an image from the `cmake-*/bin` directory that matches your build configuration.
</details>

<details>
<summary>How do I configure Ethos-U for the run, and what must match?</summary>
Set the NPU MACs using `-C mps4_board.subsystem.ethosu.num_macs` on the FVP command line. The number of MACs must match the value used when building the example.
</details>