Skip to main content

The Axelera Forum

A dedicated space for discussions, sharing ideas, connecting with other members, and about all things Axelera (besides support) — our products, innovations, company and community

  • 15 Topics
  • 26 Replies
15 Topics
Spanner
Axelera Team
SpannerAxelera Team
posted in 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? 👇

38210
Spanner
Axelera Team
S
Cadet
Sai_NikhilCadet
posted in The Axelera Forum
Export YOLOv8s Detections to JSON with Timestamps in Voyager SDK

I’m using Axelera’s Voyager SDK to run a YOLOv8s object-detection pipeline on video files, and I need to capture every detection into a JSON array for downstream processing. For each detected object, I want to record: frame_index: the zero-based frame number timestamp: the frame’s PTS in seconds bbox: [x1, y1, x2, y2] coordinates class_id: numeric COCO class index class_name: readable label (e.g. “person”) score: detection confidence What’s the cleanest way to hook into the Voyager inference loop, pull out the ObjectDetectionMeta (task name “detections”), convert its boxes, scores, class_ids, and labels into native Python types, attach the FrameResult.src_timestamp, and dump the results into a single JSON file? Example snippets or pointers to the relevant SDK APIs would be greatly appreciated.

511
S
Axelera Team
S
Cadet
saitejajampulaCadet
asked in The Axelera Forum
Anyone Tried Running Axelera with ROS 2? Facing video_infer Output Issue

Hi Axelera Community,Has anyone here successfully run the Axelera Voyager SDK with ROS 2? I’m currently working on integrating the video_infer node with ROS 2 and would really appreciate hearing from anyone who has tried this setup.My Setup: Hardware: Aetina board SDK Version: voyager-sdk v1.2.5 ROS 2: Humble (on Ubuntu 22.04) Python: 3.10 (using a virtual environment) Model Used: yolov5s-v7-coco Inference Server:python3 inference_server.py yolov5s-v7-coco none --port 50051 --no-display ROS 2 Node:ros2 run axelera_infer video_infer What’s Working: The inference server starts successfully and shows: Server started, listening on 50051 INFO : Using device metis-0:1:0 The ROS2 node launches and shows: [INFO] [video_infer_node]: Starting video inference node... [INFO] [video_infer_node]: Sending inference request... [INFO] [video_infer_node]: StreamInfer started successfully No errors are thrown on either side. Problem: The stream only seems to run a few frames: Stream Playing

925
Spanner
Axelera Team
Spanner
Axelera Team
SpannerAxelera Team
posted in 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. 👍 

1060
Spanner
Axelera Team
jonask-ai
Axelera Team
jonask-aiAxelera Team
posted in 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 

720
jonask-ai
Axelera Team
A
Anonymous
posted in 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?

411
Cherryland120
Cadet
jonask-ai
Axelera Team
jonask-aiAxelera Team
published in 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!

1830
jonask-ai
Axelera Team
S
Cadet
Semih_Cadet
posted in The Axelera Forum
Legal Issues and Axelera

Hi everyone,I am Semih. I am a PhD researcher at University of Tuebingen on Open Source and AI. I am trying to understand Axelera’s hardware and in future I am planning to research legal and regulatory issues related to AI hardware, role of open source hardware, licensing contracts, regulatory compliance especially with Cyber Resilience Act, GDPR and AI Act. I was wondering if anyone is conducting a similar research or are there any pain points or interesting discussions under these themes?Thanks in advance

270
S
Cadet
S
Ensign
saravananEnsign
asked in The Axelera Forum
Deploying lane dection model (yolopv2.onnx) on metis M2

Hello,I am currently working on lane detection model (YOLOv2) which is in .pt format I have converted to ONNX format.Here are the terminal responses I have got..root@aetina:/voyager-sdk# cd YOLOPv2root@aetina:/voyager-sdk/YOLOPv2# lsLICENSE    data     demo2.py        models            runs   yolopv2-envREADME.md  demo.py  export-onnx.py  requirements.txt  utilsroot@aetina:/voyager-sdk/YOLOPv2# cd modelsroot@aetina:/voyager-sdk/YOLOPv2/models# ls__pycache__  common.py  experimental.py  yolopv2.onnx  yolopv2.ptroot@aetina:/voyager-sdk/YOLOPv2/models# I am having the ONNX format of the model so now I want to deploy the model onto Metis M2. When I am trying to compile the yolopv2.onnx model but i am encountering this issue.root@aetina:/voyager-sdk# /voyager-sdk/bin/voyager-compile models/yolopv2.onnx --output-dir build/yolopv2 --input-shape "1,3,640,640"bash: /voyager-sdk/bin/voyager-compile: No such file or directoryI request to guide me to successfully deploy the model.Note; I am using

744
S
Ensign
Axelera CMX
Axelera Team
Axelera CMXAxelera Team
posted in The Axelera Forum
We have our first Ensigns! 🥇🥈🥉 Report to the bridge! 📯

This is a quick shoutout out to ​@ChinmayMulgund and ​@oguz for becoming the first members of Axelera AI’s community to earn the rank of Ensign! And practially at the same time (close enough that I can’t tell who got there first).Every activity on the community earns you points, and helps you climb the ranks. The opening rank is a Cadet, which you receive after your first action on the community. But it takes effort to climb the ranks, so a big congratulations to our first two Ensigns for getting there so quickly!Who’s going to be the first Lieutenant, I wonder? 

270
Axelera CMX
Axelera Team
Spanner
Axelera Team
SpannerAxelera Team
posted in The Axelera Forum
Did you see the Metis Compute Board is open for pre-orders?

Heads-up if you’re building a new AI project from scratch – the new Metis Compute Board is now up for pre-order!It’s a compact SBC with the Metis AI accelerator baked right in, powered by the ARM-based RK3588. And it runs the Voyager SDK out of the box. Loads of interfaces, solid memory specs, tiny footprint (170x170mm), and very dev-friendly.Integrated Metis AIPU with 214 TOPS for multi-stream, multi-model AI workloads 16 GB RAM for the CPU + 4 GB for the AIPU, with storage and connectivity for almost any setup Comes with Voyager SDK and BSP for fast prototyping and deploymentThere’s a product brief attached if you want to see a few more details, too.Bonus: pre-orders get 15% off. Shipping is slated to begin in June 2025, and you get 15% off for pre-ordering.Check it out: https://store.axelera.ai/products/metis-compute-board-with-arm-based-rk3588 

870
Spanner
Axelera Team
C
Cadet
carloCadet
posted in The Axelera Forum
wanted: collaborators for art&ai on edge

hi there,i am currently writing up an application with friends for a grant on “ai&art” which is due by tomorrow.we want to do music reactive visuals on edge with multimodal inputs and want to use the Metis for live inference. i think of visuals being an interesting playground since they allow to very freely construct the loads for the chip. if you are interested in collaborating please feel free to reach out! there is no requirement on expertise.  ideas only are also very much welcome.(as well as additional sources for funding or sponsors)hope to have caught your interestbest carlo 

311
Spanner
Axelera Team
Spanner
Axelera Team
SpannerAxelera Team
posted in The Axelera Forum
Check out the Axelera Metis PCIe running on an Arduino Portenta X8

As part of the recent Arduino Day celebrations, our very own ​@Steven Hunsche joined with the Arduino team to demo the Metis PCIe running on a Portenta X8!This is an unmodified LLM, running in real-time, and completely offline.It doesn’t get much more edge than these two device working in combo 😄 And there’s an early access program in the works for anyone who’s interested in getting a closer look at Arduino’s and Axelera AI’s devices in action together: https://axelera.ai/development-systems/arduinoIn the meantime, check out the project on the Arduino Day live stream: 

1400
Spanner
Axelera Team
V
Axelera Team
Victor LabianAxelera Team
posted in The Axelera Forum
Early demo: Metis M.2 running YOLOv8 on a Raspberry Pi 5

Wanted to share a quick demo we’ve been playing around with — we’ve got the Metis M.2 up and running on a Raspberry Pi 5, doing real-time vehicle detection using an unmodified YOLOv8 model.Pretty smooth experience, getting them working together. Definitely still early days, so no promises yet on timelines or any kind of future support, but figured it was worth showing that it can be done!Curious — would people here be interested in standalone edge AI running on something like a Pi5? Any particular use cases that come to mind?

2234
Spanner
Axelera Team
Spanner
Axelera Team
SpannerAxelera Team
published in The Axelera Forum
Axelera Community Gamification: Points, Ranks and BadgesTutorial

🏅 What Is Gamification on the Axelera Community?Gamification is a simple idea: it turns everyday actions into opportunities to earn points, rewards, and recognition. In the Axelera AI Community, it means your contributions—big or small—don’t go unnoticed. Whether you’re asking questions, sharing answers, or reacting to posts, you’re building up your profile as an active, valued member of the edge AI community. 🥇🥈🥉 The LeaderboardThe leaderboard serves as a dynamic scoreboard, showcasing the most active and helpful members based on the points they’ve accumulated. It’s a transparent way to recognize top contributors and inspire friendly competition. Points Accumulation: As you engage in various community activities—such as posting answers, initiating discussions, or receiving likes—you earn points. Real-Time Rankings: The leaderboard updates in real-time, reflecting the current standings of members based on their point totals. This means your position can change as you and others con

1040
Spanner
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.