The efficient edge AI platform
Stay up-to-date with benchmarks of popular deep learning models on edge hardware.
- Compare models.
- Find the best model for your application by exploring on-device benchmarks.
- Compare hardware.
- Ensure your performance requirements are met before investing in hardware.
- Keep up with trends in edge AI.
- Stay ahead of the curve and build highly efficient applications.
Data you can trust
When looking for benchmark data today, you often resort to scattered sources collected with outdated software. This can be a frustrating process. We decided we could do better!
The result? Trustworthy, up-to-date data — all in one place.
- A decade's experience.
- Embedl has been building and researching resource-efficient AI with industry leaders for nearly a decade.
- Best practices.
- At Embedl Hub, we collect all the experiences we have gained and make them available to the rest of the edge AI community.
The right model for your hardware
Models with similar accuracy often have different inference latencies depending on their architecture and the processor they run on.
Build efficient edge AI applications by finding models that make the most of your hardware.
Device:
Raspberry Pi 4 Model B
Supported hardware
More and more hardware vendors are adding specialized accelerators for deep learning on their edge devices. Access benchmarks on everything from low-powered microprocessors to powerful embedded GPUs.
Empowering the edge AI community
We are providing our platform for free in the hope that it will be a powerful resource for developers who are passionate about edge AI.
Frequently asked questions
Who is Embedl?
Embedl is a deep tech company on a mission to make edge AI extremely efficient. We create model optimization tools that empower developers to build affordable, fast, and sustainable edge AI products.
We're a spin-off from Chalmers University of Technology in Sweden, and research remains a central part of what we do. We combine proprietary innovations with the latest advancements in research to develop our state of the art tools.
Can you benchmark my custom model?
Let's talk! Please and tell us more about your use case.
How do I get in touch if I have additional questions?
If you can't find the answers you're looking for, please . We're happy to answer your questions and help you get started!
This project has received funding from the European Union’s Horizon 2020 research and innovation
programme under grant agreement No 780681, No 957197 and No 190109922.