The Ara software development kit (SDK) streamlines the development and deployment of a wide range of AI models on Ara240 DNPUs and modules. The Ara SDK features an extensible AI compiler to support the diverse and evolving model architectures; including convolutional and fully connected neural networks, transformers for vision tasks, as well highly complex models for language like LLMs, as well as Vision-Language, coding and function calling (agentic) models. The SDK offers efficient dataflow optimization, flexible quantization methods and support for multiple datatypes, ensuring high-performance and low-latency for edge-AI applications.
Choose a diagram:
| Release and Documentation | Description | Build Sources | Supported Platforms/Binaries |
|---|---|---|---|
|
Ara Software Development Kit (SDK) |
Software Development Kit enabling users to quantize and compile CNN, LLMs and VLMs for running inference on Ara240 DNPU. |
For a list of validated models and supported host and hardware platforms for Ara SDK and Runtime, please see RN00543: Ara SDK Release Notes. |
|
|
Optimum-Ara |
Optimum Ara is an extension of the Hugging Face Transformers and Optimum ecosystem that adds support for the Ara240 DNPU, enabling high-performance AI inference at the edge. |
|
|
|
Ara2 Runtime Package |
Start Here - Runtime SDK for AI/ML acceleration with Ara240 DNPU on i.MX8MPLUS and i.MX95. Means to perform benchmarks with pre-compiled CNN models. |
|
The features provided by this package are now available as a part of i.MX Embedded Linux image version 6.18.20_2.0.0 Supported Platforms: |
|
Ara2 Vision Examples |
Package containing GStreamer pipeline for 8 video streams using Object Detection models on Ara240 DNPU. |
|
The features provided by this package are now available as a part of GoPoint Application Hub integrated in i.MX Embedded Linux image version 6.18.20_2.0.0 Supported Platforms: |
|
LLM Edge Studio |
Qt based launcher for unimodal LLMs running on Ara240. |
|
The features provided by this package are now available as a part of GoPoint Application Hub integrated in i.MX Embedded Linux Image version 6.18.20_2.0.0 Supported Platforms: |
|
VLM Edge Studio |
Multimodal GenAI at the Edge - Demo showing Ara240 multimodal capabilities on i.MX with VLM, for video and image analysis. |
|
The features provided by this package are now available as a part of GoPoint Application Hub integrated in i.MX Embedded Linux Image version 6.18.20_2.0.0 Supported Platforms: |
|
eIQ AAF Connector |
A service that exposes an OpenAI‑compatible REST API to client applications while brokering requests to locally hosted LLM/VLM backends on Ara240 DNPU. The connector is optimized for the Agentic AI Framework (AAF) solution on i.MX processors and Ara DNPUs. |
|
The features provided by this package are now available as a part of i.MX Embedded Linux Image version 6.18.20_2.0.0 Supported Platforms: |
|
Smart Device Gateway |
Service interface that exposes OpenAI compatible REST API to client applications while brokering requests to locally hosted LLM/VLM backends on Ara240 DNPU. |
|
The features provided by this package are now available as a part of i.MX Embedded Linux Image version 6.18.20_2.0.0 Supported Platforms: |
| Release and Documentation | Description | Build Sources | Supported Platforms/Binaries |
|---|---|---|---|
|
Ara2 Runtime Package |
Start Here - Runtime SDK for AI/ML acceleration with Ara240 DNPU on i.MX8MPLUS and i.MX95. Means to perform benchmarks with pre-compiled CNN models. |
|
Supported Platforms: |
|
Ara2 Vision Examples |
Package containing GStreamer pipeline for 8 video streams using YOLOv8 models on Ara240 DNPU |
|
Supported Platforms: |
|
LLM Edge Studio |
Qt based launcher for unimodal LLMs running on Ara240 |
|
Supported Platforms: |
|
VLM Edge Studio |
Multimodal GenAI at the Edge - Demo showing Ara240 multimodal capabilities on i.MX with VLM, for Video and Image Analysis |
|
Supported Platforms: |
|
eIQ AAF Connector |
A service that exposes an OpenAI‑compatible REST API to client applications while brokering requests to locally hosted LLM/VLM backends on Ara240 DNPU The connector is optimized for the Agentic AI Framework (AAF) solution on i.MX processors and Ara DNPUs |
|
Supported Platforms: |
1-5 of 7 downloads
Note: For better experience, software downloads are recommended on desktop.
Quick reference to our documentation types
4 documents
Compact List
4 hardware offerings
