Skip to main content

Recent Topics

24
S
Cadet
saadtiwanaCadet
 Project Challenge: Recognize & React
Sauron – The Intelligent Gimbal (Pioneer10 project)

Hey everyone,I'm Saad, and I'm thrilled to share that my project has been selected for Axelera's Pioneer 10 Project Challenge! I am a practicing engineer (EE). When I'm not at work, you can usually find me in my home lab, working on different projects. I especially love projects that interact with the real world, like those involving machine vision. What Am I Building?I'm developing an intelligent gimbal system that will act like a "Sauron eye" for the sea right outside my balcony! Yes, the evil Sauron’s eye, but I am hoping this one will be slightly less evil. This 2-axis platform will carry cameras and, powered by Axelera's AI hardware, continuously scan the waters. My system will detect and classify different types of boats and track their activity (location, time, etc) to identify potential illegal fishing. If a fishing boat is spotted in restricted zones or at the wrong times, the system will send an immediate alert via Alexa or trigger an alarm. The gimbal will also go into track

363
Spanner
Axelera Team
A
Axelera Team
A-TangAxelera Team
Simplifying Model and Pipeline Deployment with the Voyager SDKBlog

Axelera AI’s A-Tang Fan and Doug Watt explain how the Voyager SDK simplifies the complex task of deploying AI-powered video pipelines on edge devices. This blog explores how its model compiler, model builder, and pipeline builder offer flexible integration paths—whether you’re enhancing an existing VMS, building a new edge product, or optimising for performance.Doug Watt  |  Director of AI Application Engineering​​​​A-Tang Fan  |  Engineering Manager - Applications, Model Deployment Lead Many real-world vision applications require the use of multiple deep-learning models, combined in sequence or parallel to perform different tasks. For example, outputting regions of people identified by an object detector into parallel models determining age and gender. Or cascading a vehicle detector into a license plate detector, followed by an optical character recognition model to read the plate. These multiple models are usually connected together with a camera to form an end-to-end inferencing pi

3194
Mariusz
Axelera Team
Spanner
Axelera Team
SpannerAxelera Team
 The Axelera Forum
Anyone else tried running AI on an Orange Pi 5 Plus?

Have you seen that ​@Victor Labian’s successfully brought up the Voyager SDK on the Orange Pi 5 Plus with a Metis M.2 module? It took a bit of work—Ubuntu flashing, driver rebuilds, and tweaking the PCIe device tree—but it’s running, and it’s fast. 😃This setup is seriously impressive for something so compact and affordable. Inference runs smoothly, and with some tuning, it could potentially be running advanced models like YOLOv8 without trouble.It’s just an experiment and a test at this point. No big plans around the Orange Pi, but it looks like a really promising start.If you’re experimenting with AI on SBCs—or thinking about it—I’d love to hear what you’re working on. Orange Pi, Raspberry Pi, custom boards, anything edge. 👍 

80
Spanner
Axelera Team
M
Cadet
mcunhaCadet
 Project Challenge: Recognize & React
Summer Sidekick: Your essential summer companion for plants and pets (Pioneer10 project)

Hi everyone,I am Marcos, a software engineer who has worked primarily in embedded systems across South America and Europe. I am currently based in Toulouse, France.What am I building?I am building a smart summer assistant called "Summer Sidekick" that uses Vision AI to help summer travelers manage the feeding of their pets and plants while they are away. This system ensures that pets and plants are fed properly and on time, allowing users to enjoy their vacation without worry. The entire system will be composed of actuators to trigger the food and water dispensers. For pets, AI algorithms will detect the pet's presence and food and water levels. For plants, their health will be monitored, as well as ensuring their water needs are met.Why does this project matter to me?Some summer travels can turn into a nightmare when organizing care for your companions (plants and pets) while you are away. Especially this summer, I will need to travel for almost 40 days far from home, and currently, t

181
Spanner
Axelera Team
M
Cadet
moorebrett0Cadet
 Project Challenge: Recognize & React
Elderly Guardian (My Pioneer 10 Project Challenge)

Hi everyone!My name is Brett, I am a full-time single dad, assistant soccer and softball coach, and software engineer based out of Portland, Oregon.For the Project Challenge, I’m building an AI-powered safety and wellness monitoring system for seniors living independently. The Elderly Safety Guardian will use computer vision with the Metis AI Accelerator to detect falls or unusual inactivity, and then check in with the resident through Alexa, turn on lights, and alert caregivers if needed. My goal is to give families peace of mind while helping elderly loved ones stay safe at home.This project matters to me because I’ve seen how difficult it can be for families to balance a loved one’s independence with their safety, especially as they get older. I want to create a solution that bridges that gap, empowering seniors to live confidently while still having support when it’s needed most.To start, I’ll be setting up the camera hardware and building out local fall and inactivity detection wi

452
Spanner
Axelera Team
farouqfahm
Cadet
farouqfahmCadet
 Project Challenge: Recognize & React
Drain Blockage Detection & Flood Response System for Lagos, Nigeria

Hey everyone,I’m super excited to share that my project, Drain Sentinel™, has been selected for the Axcelera AI competition and I couldn’t be more fired up about what’s ahead.So, what am I building?A vision-powered AI system that watches over storm drains and detects blockages before they turn into floods. It’s like giving cities a real-time eye on their drainage system with smart alerts, voice warnings, and action triggers to keep the streets safe.This project matters to me because I’ve seen firsthand how something as simple as a blocked drain can shut down entire communities, destroy homes, and put lives at risk. We lose time, money, and momentum every time a flood hits, and it’s time we got smarter about preventing it.DrainSentinel aims to solve that by flipping the script: from reacting to disaster… to predicting and preventing it.I’ll be starting by training our AI model to recognize clogged drains, connecting it to real-time alerts, and running live tests in simulated environment

241
Spanner
Axelera Team
S
Cadet
saraCadet
 Voyager SDK
inference on a frame obtained from the memory space

Hi again ​@Spanner ! 😊I need to infer an image obtained from byte data stored in a memory space (https://docs.python.org/3/library/mmap.html) and after converting the bytes I get an image.I've seen the sources available for create_inference_stream and I know that I can make inferences to images (https://github.com/axelera-ai-hub/voyager-sdk/blob/release/v1.2.5/docs/tutorials/application.md), but I have to specify the/path/to/file. But in this case I won't have that information.In the code I'll have this line where I read the bytes in the memory space:        frame_size=image_width*image_height*3 //RGB        frame_bytes = mm.read(frame_size)And then I get the new image:      frame = np.frombuffer(frame_bytes,dtype=np.uint8).reshape((image_height,image_width, 3))Is there any way to infer “frame”?Thank you!

584
Spanner
Axelera Team
Steven Hunsche
Axelera Team
Steven HunscheAxelera Team
 LLMs, NLP, Neural Networks
Llama 3.2B Chatbot Demo on Metis - Fully Offline

I just built a quick demo showing the Llama 3.2B chatbot running on our Metis platform, totally offline. This model packs 3 billion parameters and runs smoothly on both a standard Lenovo P360 with our PCIe card and even on an Arduino-based dev board (Portenta X8).We hit 6+ tokens/sec per core – which means real-time chat. Perfect for smart customer support bot, digital concierge systems, any edge AI assistant application really, all running fully on-device. No cloud needed.Check out the video and let me know what you think. Any projects you can think of where you could use a self-contained, power-efficient, offline AI chatbot like this? //EDIT I am aware that the youtube link is currently broken. I will reupload it soon. 

2377
Spanner
Axelera Team
jonask-ai
Axelera Team
jonask-aiAxelera Team
 The Axelera Forum
🎥 Your First Inference on Metis with Voyager SDK: Step-by-Step Video Guide

Hey everyone,Excited to share a brand-new video I put together for anyone getting started with the Metis card and Voyager SDK. Whether you're setting up your eval kit or just curious about how inference works on Metis, this walkthrough is for you.🛠️ What’s inside:Setting up your environment and verifying camera input Running YOLOv5s-v7-COCO on a live camera stream Using download_prebuilt for faster setup Running inference on media files and datasets Benchmarking performance Measuring accuracy with mAP and more📺 You can find this tutorial, and more, on our YouTube ChannelThis guide is based on SDK version 1.2.5 and follows the official Quick Start Guide, but with some extra tips and tricks to get you up and running even faster.💬 Have you tried running inference on Metis yet?Drop your thoughts, questions, or your own setup experiences in the comments, I'd love to hear how it's going for you or help troubleshoot if you're stuck!See you there,Jonas 

290
jonask-ai
Axelera Team
I
Cadet
IAKCadet
 Metis M.2
Problems with network inference. Uio wait kernel failed with return code -1406.

Hello!I am facing some problems tring to run a network inference on the Metis M.2 with Raspberry Pi via Docker. To try the inference I have followed the steps in the guide (https://support.axelera.ai/hc/en-us/articles/26362016484114-Bring-up-Voyager-SDK-in-Raspberry-Pi-5) but after trying to run the inference.py, I face the following errors:  (venv) root@iak:/home/voyager-sdk# ./inference.py yolov8s-coco-onnx ./media/traffic1_480p.mp4INFO    : Using device metis-1:1:0WARNING : Failed to get OpenCL platforms : clGetPlatformIDs failed: PLATFORM_NOT_FOUND_KHRWARNING : Please check the documentation for installation instructionsINFO    : Default OpenGL backend gl,3,3 overridden, using gles,3,1INFO    : Network type: NetworkType.SINGLE_MODELINFO    : InputINFO    :   └─detectionsStream Paused:  57%|██████████████████████████████████████████████████████▊                                         | 4/7 [00:00<00:00,  5.56/s][ERROR][axeWaitForCommandList]: Uio wait kernel failed with return c

12417
V
Axelera Team
M
Cadet
maximiliankirCadet
 Voyager SDK
Yolo Performance Issues

Hello, I try to use a YOLO model on 8 camera streams in parallel on the Metis PCIe card. 20-30 fps per camera stream is all I need. With the 548 fps end-to-end stated for YOLOv8s in https://axelera.ai/metis-aipu-benchmarks, it should be possible to reach ~68 fps per camera. As a small test I wrote a python script, which creates a inference stream with 8 videos as input:from axelera.app import config, display, inf_tracersfrom axelera.app.stream import create_inference_streamdef run(window, stream): for frame_result in stream: window.show(frame_result.image, frame_result.meta, frame_result.stream_id) fps = stream.get_all_metrics()['end_to_end_fps'] print(fps.value)def main(): tracers = inf_tracers.create_tracers('core_temp', 'end_to_end_fps', 'cpu_usage') stream = create_inference_stream( network="yolov5s-v7-coco", sources=[ str(config.env.framework / "media/traffic1_1080p.mp4"), str(config.env.framework / "media/traffic1_

775
M
Cadet
S
Cadet
sifat0433Cadet
 Metis M.2
Input/Output Error Running Inference with Metis M.2 on ASRock Z170 Pro4S

Hello everyone,I’m new to working with this hardware and recently installed the Metis M.2 chip (Firmware version: v1.2.0-rc2+bl1-stage0) on an ASRock Z170 Pro4S motherboard. The device was detected successfully.I’ve installed the Voyager SDK and attempted to run the following command from the Quick Start Guide:./inference.py yolov5s-v7-coco dataset --no-display However, I encountered an input/output error when executing the command. I’ve included the full error message below for reference.I would appreciate any guidance or suggestions you may have to help resolve this issue.Thank you in advance for your support. INFO    : Using default val dataset INFO    : Using device metis-0:5:0 INFO    : Network type: NetworkType.SINGLE_MODEL INFO    : Input INFO    :   └─detections INFO    : Imported DataAdapter ObjDataAdaptor from /home/tud/voyager-sdk/ax_datasets/objdataadapter.py Stream Playing:  71%|██████████████████████▏        | 5/7 [00:00<00:00,  6.26/s][AxeleraDmaBuf.cpp:234] UIO_IOCTL

775
S
Cadet
A
Anonymous
 The Axelera Forum
Which demos would best showcase the strengths of our product?

We frequently attend trade shows and aim to stand out from the competition. Showcasing compelling demos is a key strategy to achieve this. Which demos would most effectively highlight the unique strengths of our product?

381
Cherryland120
Cadet
Spanner
Axelera Team
SpannerAxelera Team
 The Axelera Forum
👋 Say Hello – Introduce Yourself to the Axelera AI Community

Welcome to the Axelera AI Community – we’re so glad you’re here.This is a space for everyone from hardware engineers and AI devs to makers, enthusiasts, onlookers and partners, so don’t be shy. Whether you’re working with AI every day or just getting started, we’d love to know more about you.To get things going, let’s have a big round of hellos:Who you are and what you’re working on Your experience or interest in the AI/edge AI world What you’re hoping to learn, share or achieve in this communityCan’t wait to get to know you all, and to build something great together.Now – who’s going first? 👇

34210
Spanner
Axelera Team
S
Cadet
saraCadet
 Voyager SDK
Raspberry Pi 5 - Metis inference with GStreamer Pipeline

Hi, I'm working on an RPI5 and I have the voyager SDK inside a container with ubuntu 22 and it works. I wanted to try running an inference on a stream using GStreamer. First, I tried to access the stream and I managed with the following command:GST_PLUGIN_PATH=/voyager-sdk/operators/lib AXELERA_DEVICE_DIR=../opt/axelera/device-1.2.5-1/omega LD_LIBRARY_PATH=/voyager-sdk/operators/lib:/opt/axelera/runtime-1.2.5-1/lib:$LD_LIBRARY_PATH gst-launch-1.0 rtspsrc location='rtsp://…..' latency=200 ! rtph264depay ! avdec_h264 ! videoconvert ! autovideosink Now I wanted to try the axinferencenet plugin but I'm having problems. I've tested various commands and formats, such as: GST_PLUGIN_PATH=/voyager-sdk/operators/lib AXELERA_DEVICE_DIR=../opt/axelera/device-1.2.5-1/omega LD_LIBRARY_PATH=/voyager-sdk/operators/lib:/opt/axelera/runtime-1.2.5-1/lib:$LD_LIBRARY_PATH gst-launch-1.0 rtspsrc location='rtsp://...' latency=200 ! rtph264depay ! avdec_h264 ! videoconvert ! video/x-raw,format=BGRA !  axtran

916
H
Axelera Team
Y
Cadet
yairc2223Cadet
 Metis PCIe
axdevice --refresh fails

0000:01:00.0 : Axelera AI Metis AIPU (rev 02)ERROR:axelera.runtime:No AIPU driver found in lsmod outputERROR: AXR_ERROR_CONNECTION_ERROR: No AIPU driver found in lsmod output 

775
V
Axelera Team
C
Ensign
ChinmayMulgundEnsign
 Metis M.2
Guidance on Running Lane Detection Model (culane_res18.onnx) on Metis M2

Hello,I am currently working with a lane detection model named culane_res18.onnx, and I would like to know how I can run this model on the Metis M2 device.Here are some details about the setup:The model is culane_res18.onnx, which is a lane detection model. The training dataset is from the CULane dataset, and I have the link to the dataset here. The model file can be found here.Could you kindly guide me on how to run this lane detection model on Metis M2? I would appreciate any step-by-step instructions or basic guidance on how to deploy this model, considering that I have both the ONNX model and the dataset.Looking forward to your response!Best regards,Chinmay Mulgund

33524
H
Axelera Team
K
Cadet
ku15921Cadet
 Metis M.2
Error installing metis-dkms

Hi,I'm having trouble installing the metis-dkms driver. Can you help me?The device is a NanoPC-T6 from Friendly Elec, and the operating system is Ubuntu 22.04, also provided by Friendly Elec.I'm leaving the error message and OS information below. If there is any more information I should provide, please let me know. Thank you. The Error:$ sudo dpkg -i metis-dkms_0.07.16_all.deb(Reading database ... 155441 files and directories currently installed.)Preparing to unpack metis-dkms_0.07.16_all.deb ...Deleting module metis-0.07.16 completely from the DKMS tree.Unpacking metis-dkms (0.07.16) over (0.07.16) ...Setting up metis-dkms (0.07.16) ...Loading new metis-0.07.16 DKMS files...Building for 6.1.99Building for architecture aarch64Building initial module for 6.1.99ERROR (dkms apport): kernel package linux-headers-6.1.99 is not supportedError! Bad return status for module build on kernel: 6.1.99 (aarch64)Consult /var/lib/dkms/metis/0.07.16/build/make.log for more information.dpkg: error pro

1015
V
Axelera Team
S
Cadet
saraCadet
 Metis M.2
Ubuntu 24.04 support for RPi5, no container

Hello!!Do you have any news about support for Ubuntu 24.04? I wanted to use axelera on raspberry pi cm5 however there is no support for 22.04 and I wanted to run it on the host, not in a docker container.Thank you 😊

201
Spanner
Axelera Team
C
Ensign
ChinmayMulgundEnsign
 Metis M.2
AXELERA M2 Not Detected – “No target device found in lspci output”

Hello,I’m having trouble getting my AXELERA M2 to be recognized under Linux. Until recently everything was working fine; now when I try to run an inference the SDK can’t find the board. I’ve already tried rebooting, but the issue persists.aravind@aravind-H610M-H-V2:~/Desktop/voyager-sdk$ source venv/bin/activate(venv) aravind@aravind-H610M-H-V2:~/Desktop/voyager-sdk$ ./inference.py yolov5s-v7-coco chinmay/Vehicle-CV-ADAS/assets/challenge_video.mp4 --save-output ~/Desktop/voyager-sdk/chinmay/saved_videos/yolov5s-v7-coco.mp4ERROR : No target device found in lspci outputERROR : AXR_ERROR_CONNECTION_ERROR: No target device found in lspci output 

623
Spanner
Axelera Team
F
Cadet
floritCadet
 
we need more DRAM … from 1TB up to 4TBNew

Hello I got some Suggestion for the next Generation metis pcie x16 4.0 cardsWe Need more DRAM E.g. 1TB/2TB/4TBUp to 8x Quad Core processorsThis will be a good goal for Future cards

2
261
F
Cadet
M
Axelera Team
Manuel MohrAxelera Team
Using oneAPI Construction Kit to Enable Open Standards Programming for the Metis AIPUBlog

Manuel Mohr | Staff Software Engineer at AXELERA AI  Open standards enable developers to more easily harness the power of AI accelerators, especially in heterogenous computing. Here you can read in detail why and how we implemented OpenCL using the oneAPI construction kit on Metis.The Necessity of Dedicated AI Hardware AcceleratorsAI applications have an endless hunger for computational power. Currently, increasing the sizes of the models and cranking up the number of parameters has not quite yet reached the point of diminishing returns. Thus, the ever growing models still yield better performance than their predecessors.At the same time, new areas for application of AI tools are explored and discovered almost daily. Hence, building dedicated AI hardware accelerators is extremely attractive. In some situations it is even a necessity, as it enables running more powerful AI applications while using less energy on cheaper hardware. Welcome to the Hardware JungleSuch specialized accelerato

21010
Spanner
Axelera Team
Spanner
Axelera Team
SpannerAxelera Team
 Project Challenge: Recognize & React
Getting started with your project build

Quickly wanted to share this excellent post by ​@jonask-ai, which will be super useful for getting started on your project straight away if it gets selected:More tutorials to come too, but this will get you up and running with the Aetina system right out of the gate.

930
Spanner
Axelera Team
jonask-ai
Axelera Team
jonask-aiAxelera Team
 The Axelera Forum
Getting Started with Voyager SDK – A Quick IntroductionTutorial

Hi everyone,I’ve just published a quick walkthrough of the Voyager SDK, our all-in-one platform for building and deploying AI applications at the edge. If you’re working on computer vision tasks or just curious about what makes Metis and Voyager such a powerful combo, this video is for you.In it, we’ll cover the basics:🔹 What Voyager SDK is and how it simplifies AI development🔹 How to use the Model Zoo or bring your own model🔹 Performance, quantization, pipelines, and real-time inference🔹 A peek at cascading models and what’s possible with our quad-core Metis AIPUWhether you’re building a demo, proof of concept, or production deployment, this will help you get started with confidence. Let me know what you think!Looking forward to seeing what you build with it!

1340
jonask-ai
Axelera Team

Badge Winners

Show all badges
Code of Conduct

Address

HTC5, High Tech Campus
5656 AE Eindhoven
The Netherlands
Email: info@axelera.ai

Menu

AI Accelerators

AI Software

Metis Evaluation Systems

Partners

Product Inquiry

Company

Our Story

Our Team

Careers

News

Contact

Reducing CO₂ with
Axelera’s Forest

Follow Us

LinkedIn

Twitter

Sign Up for Our Newsletter

EU flag Axelera AI is co-funded by the European Union’s Horizon research and innovation programme.

Privacy Policy  |  Cookie Policy  |  Terms of Sale  |  Vulnerability Disclosure Policy

Copyright 2025 © All rights Reserved.