> For the complete documentation index, see [llms.txt](https://nx.docs.scailable.net/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://nx.docs.scailable.net/ai-manager-v6.1.3/ai-accelerators-support/introduction-to-oaxs.md).

# Accelerators Overview

The AI Manager accelerates AI models inference using various built-in **runtimes**, each of which is dedicated to a specific platform and AI accelerator (CPU, GPU, NPU, etc.). These runtimes are seamlessly integrated within the AI Manager, facilitating effortless utilization for individuals employing new runtimes and harnessing various forms of AI acceleration.

When a user uploads a [supported](/ai-manager-v6.1.3/ai-models-support/importing-models.md) AI model to the AI Cloud, the uploaded model file is subject to several conversion processes, **each of which generates a new model artifact that is used by one of the** [**runtimes**](/ai-manager-v6.1.3/ai-accelerators-support/supported-ai-accelerators.md) **that are provided.**

### Example of a Teachable Machine model

Suppose you've trained a teachable machine model, and exported it according to the guidelines mentioned [here](/ai-manager-v6.1.3/ai-models-support/importing-models/from-teachable-machine.md), the following model artifacts are stored to be used by the AI Manager when needed:

<figure><img src="/files/0TJ19VPw0zBHpLVEmPLZ" alt="Model detail page showing available artifact download types: Default ONNX, Original ZIP,
     Hailo 8L ONNX, and Hailo 8 ONNX"><figcaption><p>Model artifact types: each runtime uses a different artifact generated from the uploaded model.</p></figcaption></figure>

* `application/zip; source=original`: is the original ZIP archive that's uploaded in the interface.
* `application/zip; kind=teachable-machine`: is the original ZIP archive that's uploaded in the interface with specific details for Teachable Machine.
* `application/x-onnx`: the TFLite generated in an earlier stage is converted to ONNX, validated and optimized to run on both **CPU**, **Intel** and **Nvidia** hardware.\
  \&#xNAN;*Please note, that in this step, no quantization is performed on the model.*
* `application/zip; device=mxa`: is the artifact generated by compiling the ONNX file into an optimized file dedicated only for MemryX hardware.
* `application/x-onnx; device=hailo`: is a custom ONNX generated specifically for Hailo-8 chips.

{% hint style="info" %}
Any runtime/toolchain combination adhering to the [Open AI Accelerator eXchange (OAAX)](https://www.oaax.org/) standard is compatible with the AI Manager, enabling straightforward substitution of any existing installed runtime with new one.\
If you're a AI chip maker and would like to integrate with the AI Manager, please refer to this documentation repository on [Github](https://github.com/OAAX-standard/OAAX).
{% endhint %}