At DeGirum, we’re committed to providing cutting-edge AI solutions that are versatile and adaptable. One of our standout features is our hardware-agnostic PySDK, designed to seamlessly integrate with various AI accelerators and application processors, allowing you to choose the best hardware for your specific needs.
| Supported Hardware | On the Roadmap | Will Be Added On Demand |
|---|---|---|
| DeGirum – DeGirum Orca | Qualcomm - Qualcomm® Snapdragon™ Neural Processing Engine (NPE) | Tenstorrent - Tenstorrent Grayskull chip |
| Intel – Intel CPUs/GPUs/NPUs | Texas Instruments - Texas Instruments TDA4VM processor, Jacinto 7 series | SiMa.ai - SiMa.ai MLSoC™ |
| NVIDIA – Jetson and Orin series (e.g., Jetson Nano, Jetson Xavier NX, Orin Nano, Orin NX, Orin AGX) | Renesas - Renesas RZ/G2 AI MPU, Renesas DRP-AI | Hailo - Hailo-8 AI Processor |
| NXP – NXP i.MX 8M Plus, NXP i.MX 93 | STMicroelectronics - STM32 series with AI extensions, STM32MP1 | Kinara - Kinara Ara-1 Edge AI Processor |
| AMD – AMD Ryzen Embedded series with integrated GPU | – | Untether AI - Untether AI runAI200 device |
| Raspberry Pi – Raspberry Pi 4/5 with AI accelerators like the Google Coral USB Accelerator | – | Axelera AI - Axelera AI Metis™ AI Platform |
| Google – Google Coral (Edge TPU) | – | Blaize - Blaize Pathfinder P1600 SoM, Xplorer Accelerator |
| Rockchip – RK3588, RK3568, RK3566 | – | Kneron - Kneron KL520 AI SoC, Kneron KL720 AI SoC |
| – | – | MemryX - MemryX MX3 AI Accelerator |
What Does Hardware-Agnostic Mean?
Our PySDK isn’t tied to any specific hardware. Whether you’re using our DeGirum Orca, a different accelerator, or running your models in the AI Hub, our PySDK ensures compatibility and flexibility. This approach gives you the freedom to build and deploy AI applications without being locked into a single hardware platform.
Why It Matters
-
Flexibility: Use the same codebase across multiple hardware platforms.
-
Scalability: Easily switch or upgrade hardware without rewriting your AI models.
-
Efficiency: Choose the most suitable hardware for each task, optimizing performance and cost.
Learn More
Explore our PySDK Documentation to see how easy it is to get started with our hardware-agnostic AI tools.
Join the Discussion
We’d love to hear how you’re using PySDK in your projects. Share your experiences and ask questions in our Community Discussions.