Hailo-10 M.2 Key-M module brings Generative AI to the edge with up to 40 TOPS of performance

Hailo-10 M.2 module generative AI for the edge

Hailo-10 is a new M.2 Key-M module that brings Generative AI  capabilities to the edge with up to 40 TOPS of performance at low power. It targets AI PCs supporting only the Windows 11 operating system on x86 or Aarch64 targets at this time. Hailo claims the Hailo-10 is faster and more energy efficient than integrated neural processing unit (NPU) solutions found in Intel SoCs and delivers at least twice the performance at half the power of Intel’s Core Ultra “AI Boost” NPU. Hailo-10 module specifications: AI accelerator – Hailo-10H System Memory – 8GB LPDDR4 on module Host interface – 4-lane PCIe Gen 3 Power consumption – Less than 3.5W (typical) for the chip Form factor – M.2 Key M 2242 / 2280 Supported AI frameworks – TensorFlow, TensorFlow Lite, Keras, PyTorch & ONNX The Hailo-10 can run Llama2-7B with up to 10 tokens per second (TPS) at under 5W […]

DeGirum ORCA M.2 and USB Edge AI accelerators support Tensorflow Lite and ONNX model formats

Degirum ORCA M.2 PCIe module

I’ve just come across an Atom-based Edge AI server offered with a range of AI accelerator modules namely the Hailo-8, Blaize P1600, Degirum ORCA, and MemryX MX3. I had never heard about the last two, and we may cover the MemryX module a little later, but today, I’ll have a closer at the Degirum ORCA chip and M.2 PCIe module. The DeGirum ORCA is offered as an ASIC, an M.2 2242 or 2280 PCIe module, or (soon) a USB module and supports TensorFlow Lite and ONNX model formats and INT8 and Float32 ML precision. They were announced in September 2023, and have already been tested in a range of mini PCs and embedded box PCs from Intel (NUC), AAEON, GIGABYTE, BESSTAR, and Seeed Studio (reComputer). DeGirum ORCA specifications: Supported ML Model Formats – ONNX, TFLite Supported ML Model Precision – Float32, Int8 DRAM Interface – Optional 1GB, 2GB, or 4GB […]

Avaota A1 open-source hardware SBC is powered by Allwinner T527 octa-core Cortex-A55 SoC

Avaota A1 open-source hardware Allwinner T527 SBC

We’ve recently covered MYiR Tech MYD-LT527 industrial development board based on Allwinner T527 octa-core Cortex-A55 AI SoC and noted Orange Pi is working on one that should even get mainline Linux support. The Avaoto A1 offers another Allwinner T527 hardware option with an SBC design that’s fully open-source. The board is equipped with up to 4GB RAM, 128GB eMMC flash, HDMI and DisplayPort video outputs, two gigabit Ethernet ports, a WiFi 6 and Bluetooth 5.4 module, a few USB ports, a 3.5mm audio jack and the usual 40-pin GPIO header for expansion. Avaota A1 specifications: SoC – Allwinner T527 (or Allwinner A527 with Avaota A1C board, not sure what the differences are between the two) CPU Octa-core Arm Cortex-A55 processor with four cores @ 1.80 GHz and four cores @ 1.42GHz XuanTie E906 RISC-V core up to 200 MHz GPU – Arm Mali-G57 MC1 GPU with support for OpenGL ES […]

SolidRun launches Hailo-15 SOM with up to 20 TOPS AI vision processor

SolidRun Hailo 15H Powered Edge AI System on Module

In March last year, we saw Hailo introduce their quad-core Cortex-A53-based Hailo-15 AI Vision processor. The processor features an advanced computer vision engine and can deliver up to 20 TOPS of processing power. However, after that initial release, we didn’t find it in any commercial products with the SoC. But in a recent development, SolidRun has released a SOM that not only features the Hailo-15 SoC but also integrates up to 8GB LPDDR4 RAM and 256GB eMMC storage along with dual camera support with H.265/4 Video Encoder. This is not the first SOM that SolidRun has released. Previously, we wrote about the SolidRun RZ/G2LC SOM, and before that, SolidRun launched the LX2-Lite SOM along with the ClearFog LX2-Lite dev board. Last month, they released their first COM Express module based on the Ryzen V3000 Series APU. Specification of SolidRun’s Hailo-15 SOM: SoC – Hailo-15 with 4 x Cortex A53 @ 1.3GHz; […]

Orange Pi Developer Conference 2024, upcoming Orange Pi SBCs and products

Orange Pi Developer Conference 2024

Orange Pi held a Developer Conference on March 24, 2024, in Shenzhen, China, and while I could not make it, the company provided photos of the event where people discussed upcoming boards and products, as well as software support for the Orange Pi SBCs. So I’ll go through some of the photos to check out what was discussed and what’s coming. While Orange Pi is mostly known for its development boards the company has also been working on consumer products including the Orange Health Watch D Pro and the OrangePi Neo handheld console. The Orange Pi Watch D Pro is said to implement non-invasive blood glucose monitoring, blood pressure monitoring, one-click “micro-physical examination” and other functions to to help users monitor their health monitoring. The Watch D Pro uses a technique that emits a green light to measure glucose levels in the blood, and we’re told it’s accurate enough to […]

BeagleY-AI SBC features TI AM67A vision processor with 4 TOPS AI accelerators

Texas Instruments AM67A single board computer

The BeagleBoard.org Foundation’s BeagleY-AI is an open-source hardware, credit card-sized SBC powered by a Texas Instruments AM67A quad-core Cortex-A53 vision processor with various programmable blocks capable of delivering up to 4 TOPS for AI algorithms. The board ships with 4GB RAM, relies on a microSD card slot for storage and OS booting, implements gigabit Ethernet, WiFi 6, and Bluetooth 5.4 connectivity, and can drive up to three displays via micro HDMI, OLDI (LVDS), and MIPI DSI interfaces. The BeagleY-AI also comes with two MIPI CSI camera interfaces, four USB 3.0 ports, a USB Type-C port, and a 40-pin GPIO header for expansion. We can also see a 16-pin PCIe FPC connector that looks somewhat similar to the 20-pin PCIe connector on the Raspberry Pi 5 but officially supports PCIe Gen3 x1. BeagleY-AI specifications: SoC – Texas Instruments AM67A (J722S) “vision processor” CPU Quad-core 64-bit Arm Cortex-A53 processor at 1.4GHz Arm […]

Ambiq Apollo510 Arm Cortex-M55 MCU delivers up to 30x better power efficiency for AI/ML workloads

Ambiq Apollo Cortex-M55 AI microcontroller

Ambiq Apollo510 Arm Cortex-M55 microcontroller delivers 30 times better power efficiency than typical Cortex-M4 designs and 10 times the performance of the Apollo4 Cortex-M4 sub-threshold microcontroller for AI and ML workloads. The new MCU also comes with 4MB NVM, 3.75MB SRAM, a 2.5D GPU with vector graphics acceleration that’s 3.5 times faster than the Apollo4 Plus, and support for low-power Memory-in-Pixel (MiP) displays. Like all other Ambiq microcontrollers, the Apollo510 operates at sub-threshold voltage for ultra-low power consumption and implements security with the company’s secureSPOT platform with Arm TrustZone technology. Ambiq Apollo510 specifications: MCU Core – Arm Cortex-M55 core up to 250 MHz with Arm Helium MVE, Arm TrustZone, FPU, MPU, 64KB I-cache, 64KB D-cache, 256KB I-TCM (Tighly Coupled Memory), 256KB D-TCM, Graphics – 2.5D GPU clocked at 96 MHz or 250 MHz with vector graphics acceleration, anti-aliasing hardware acceleration,  rasterizer/full alpha blending/texture mapping, texture/framebuffer compression (TSC4, 6, 6A and […]

MaaXBoard OSM93 – Business card-sized SBC features NXP i.MX 93 AI SoC, supports Raspberry Pi HATs

MaaXBoard OSM93 SBC

MaaXBoard OSM93 is a single board computer (SBC) based on a Size-S OSM module powered by an NXP i.MX 93 Cortex-M55/M33 AI SoC and offered in a business card form factor with support for Raspberry Pi HAT boards through a 40-pin GPIO header and mounting holes. The board also comes with 2GB LDDR4, 16GB eMMC flash, MIPI CSI and DSI interfaces for optional camera and display modules, two gigabit Ethernet ports, optional support for WiFi 6, Bluetooth 5.3, and 802.15.4, three USB 2.0 ports, and two CAN FD interfaces with on-board transceivers. MaaXBoard OSM93 specifications: SoC – NXP i.MX93 CPU 2x Arm Cortex-A55 up to 1.7 GHz 2x Arm Cortex-M33 up to 250 MHz GPU – 2D GPU with blending/composition, resize, color space conversion NPU – 1x Arm Ethos-U65 NPU @ 1 GHz up to 0.5 TOPS Memory – 640 KB OCRAM w/ ECC Security – EdgeLock Secure Enclave System […]

UP 7000 x86 SBC