Embedded Learning Library (ELL)
FreeMicrosoft's library to deploy intelligent machine-learned models onto resource constrained platforms and small single-board computers.
FreeFree tier
About Embedded Learning Library (ELL)
The Embedded Learning Library (ELL) is an open-source library from Microsoft Research that enables the design and deployment of intelligent machine-learned models onto resource-constrained platforms and small single-board computers such as Raspberry Pi, Arduino, and micro:bit. Deployed models run entirely locally, requiring no network connection or cloud servers. ELL is an early preview and was archived on July 17, 2024; it includes tutorials, a website with pretrained models, and installation guides for Windows, Linux, and macOS.
Key Features
Design and deploy machine-learned models to resource-constrained devices
Supports Raspberry Pi, Arduino, and micro:bit
Local inference without network connection or cloud servers
Includes tutorials, installation guides, and a gallery of pretrained models
Developed at Microsoft Research as an early preview
Open source under MIT license
Pros & Cons
Pros
- Enables local, offline machine learning inference on low-power hardware
- Open source with MIT license
- Backed by Microsoft Research
- Supports multiple popular single-board platforms
- Includes pretrained model gallery and comprehensive documentation
Cons
- Repository archived and no longer actively maintained (read-only since July 2024)
- Early preview with frequent breaking API changes
- Limited to supported hardware (Raspberry Pi, Arduino, micro:bit)
- Development community has moved on; no further updates expected
Best For
Deploying machine learning models on edge devices like Raspberry Pi and ArduinoIoT and embedded systems requiring offline inferenceEducational projects with micro:bit and single-board computersPrototyping local AI applications without cloud dependencies
FAQ
What is the Embedded Learning Library (ELL)?
ELL is an open-source library from Microsoft Research that allows you to design and deploy machine-learned models onto resource-constrained platforms like Raspberry Pi, Arduino, and micro:bit, running locally without cloud dependencies.
Is ELL still actively maintained?
No, the repository was archived by Microsoft on July 17, 2024, and is now read-only. It is no longer actively developed or maintained.
What hardware does ELL support?
ELL targets small single-board computers such as Raspberry Pi, Arduino, and micro:bit.
Does ELL require an internet connection?
No, deployed models run entirely locally and do not require a network connection or cloud servers.
Where can I find tutorials and pretrained models?
The official ELL website at microsoft.github.io/ELL provides tutorials, instructions, and a gallery of pretrained models.