Seeed Studio’s Post

View organization page for Seeed Studio, graphic

42,733 followers

2mo Edited

😎Curious about our hottest #reComputer J4012 edge device? Join us as we unbox the AI computer, power it up, and explore its incredible capabilities in computer vision and large language models for diverse applications, such as robotics, automation, video analytics, and Generative AI. Watch the video on YouTube: https://lnkd.in/gXfiMpVP What you'll discover: ✅ reComputer part list ✅ Take apart the reComputer to check out the specifications breakdown of J401 carrier board ✅ Plug in to complete the initial setup ✅ Configurations for installed packages and runtime environment ✅ One-line command deployment of Ultralytics Yolov8 and Meta Llama3 on reComputer Ready to kickstart your Edge AI Project? Check out reComputer J4012, powered by NVIDIA Robotics Jetson Orin NX 16GB for infinite possibilities! https://lnkd.in/g7J7w7Ts Explore more LLM and CV models supported on Jetson Examples GitHub: https://lnkd.in/gPcgtSuq #nvidia #jetson #edgeai #computervision #videoanalytics #llm #generativeai #robotics #autonomous

6 Comments

Transcript

Hi there, welcome. This is Studio AJ channel. Today we're going to unbox the recomputed J 401 two, our 23 customer top all day rate things device from the Jason Howard product line. We check out what's inside the package, how to set it up as a rectangle computer and how to deploy the computer vision and local large language models on it with simple steps. After watching this video, you can get started with the device. Is it easy? The Recomputed J4012 is a powerful and compact embedded Agii computer powered by the combination of the open source cardboard and the Media Jensen Orianna X16GB production module, which can deliver up to 100 TOPS. They have performance by combining the 1024 core NVIDIA Ampere architecture GPU with 32 Tensor cores and the 8 core 64 bit CPU for powerful operating capability we've pre installed. They've had five point 1.1 on system with various certifications to help accelerate your product to market. With great support from Oxen and Bologna. You can also deploy your devices in large scale or convenient for the device management and BSP OTA updates. He made diverse application and influencing power requirements with extend The recomputed as a whole product series covered full range of Jason Mojos from Jason Nano X to Oriano. And relax, providing aid performance from 0.5 Tflops to 100 tops. Here's how to do it instead of by cleaning your Easter space, then gathering all the parts from all the package. That includes the recompute, the age box and the power adapter as 90 watts. 5 amber standard. Now let's take a closer look at the recomputed J4012O device. Here's the classic studio logo either front panel. We also provide logo customization for a solution in production. Sad Piano is surrounded by aluminum frame and for the IO panel we provide convenient and practical connectors including the DC power Jack, HDMI 2.14, USB 3.2 Type A, RJ45, Gigabit Ethernet and USB 2.0 Type C. The upper cover is magnetically connected to the whole case so that you can easily open the top cover while protecting the integrity of the H device without losing or damaging the internal components. On the backside you'll see there are four screw holes in case you need to mount your recompute on the wall or place at the desktop computer Is actually a hand size measured of 38 centimeter less 12 centimeter wise and 5.85cm height. Which can be easily baited into various scenarios from autonomous machine to industrial system. We also provide comprehensive supplications such as roles EFCC, KC, UTCA ensuring your product to market. Besides the basic connections, let's look inside what's other highlight features? There's an open source where computer J401, Carrier Board AJ and Relax Mojo connected with the fan. Now let's take apart these components to get the full picture of the keyboard. Well, doesn't matter for this part. We use two types of screwdrivers, the 2.5mm crowd screwdriver and the 2.5mm Allen wrench as Hashagen. There are four schools that host the board with the case. Now let's remove these screws up. And now we can remove the camera board. Remember also remove the fan connector from the four P PWM fan interface. The jazz emoji secured with the board or other two schools. After removing them, there are two remaining clips on the side of the Mojo. So we should release the clips to slide up the mojo from the 216 pin solemn factor. Besides the interfaces that we mentioned before, they are also due CSI camera connectors. UART hider Ken. Ford being GPIO on the underside of the board we have the m. twokeym slot for extra NVM express SSD connection. Here we provide a 128 GSD with Jetpack 5 point 1.13 start on it so that you can directly power on the device without flashing configurations. There are also under two key E for wireless connection with Wi-Fi or Bluetooth modules. And here's the RTC socket. Now let's raise the recommended J40 into the case quickly example all parts together and power it on for the next development. Let's connect. All the necessary equipments. Here's the Ethernet, mouse and keyboard display. And finally, plug the power in. Notice that the part list doesn't include the power cable, since the specifications for power cables vary from country to country for customers around the world. So please choose the appropriate power cord module for your country when you purchase on the website, whether it's UIS or European standard. Once the green light shows up, it's. Would you go? The initial setup takes a couple of minutes with preferred language setting username and password. This kind of usual questions the log in to get into the main workspace. To make sure we have sufficient packages available for our further AI deployment, we should check in the terminal to install the necessary package first. Now we install the Jet Top tool to monitor usage of hardware and also the MVP model to release all JSON resources to the Max. Then they report that device since every computer has pre-installed JPEG 5.1 point one version before delivery, so it's already integrated with CUDA environment, OpenCV deepstream packages and etc. Now we should configure the Docker runtime environment. Here you can refer to the NVIDIA official tutorial. Now the device is ready for the AI project. To quickly get experience with the Jason development and deployment, we provide the open source Jason example GitHub repository with various computer vision and large language model applications to help simplify your deploying pipeline. Here we collect each example in a Docker image. You only need to use this one line command to deploy on your local site. Let's take an example of this ultra latest euro. Each docker page you'll learn what's the model used for, how to download and run it with simple steps, and how the result look like. Now we follow the quick start guide in Terminal install Jason example packages. Then we check the list of models that J examples currently supported. Here we find outlet's Europe and we run it on Jason with one line comment. There were moments when you first load it. Once you see the IP address, the download part is completely successfully. Now let's go to the web UI to run mode on Jackson. Here you can choose the model size and also the input type as live stream or image or video and upload your input source This directly. Here's loading around one minute and Della. Here you'll see the detection result. On the right side you accelerate the inferencing speed. For better performance you should convert model from the PT to the tensor D format. You can follow Vt to quickly learn how to use tensor RT. And deep strain mountains. More than that, you can also one command deploy such as Metal Lama 3 to get more human like response efficiently. Now let's chat with the virtual assistant. You can see that the model can be loaded normally because there's only half memory use currently and the model is totally running on the GPU with great token generation speed. Now feel free to explore your AGI world if you're looking for a powerful embedded age AI device to get experience after latest Vision AI and Liam technologies. That's the one with the strong inferencing capability and the latest Jetpack 6.0 support. It really helps simplify the HMI development and deployment. All right, last but not least, if you find this way to useful, don't forget to like the video and subscribe our channel to get more latest HIV. Stars updates label comment anytime when you meet any problems by using the recomputed JSON device. We're always here to help. See you next time.

I AM HERE, COUNT ON ME! Interested to support digital transformational experiences (focus on AIOT)! Automotive Project/Program Manager with focus on Cybersecurity (ASPICE, ISO21434) and Functional Safety/FuSi (ISO26262).

2mo

I just deployed this J4012 with a project for a customer which has an egg farm. It was not an easy task because before Jetpack 6, it was quite hard to make all updates. I would love to share a blogpost with you about what we did etc. and also how it works. we used thus non-industrial model, but I recommend also the reserver Industrial and maybe I will give a try to reComputer Industrial since I have some ideas about what we can do to improve.

AI Engineer

2mo

Orin NX with 100 TOPS is impressive, especially compared to Xavier NX with only 21 TOPS. Jetson Orin NX 16GB does not have EMMC on its SOM, and on the reComputer, SSD is used.

Arthur Zarankin 💡

Arthur Zarankin 💡

💻 Software Engineer 📎 AI, HPC Consultant 📎 C++, CUDA® Developer

2mo

Does it stand for industrial standards or is this kit for development purposes?

See more comments

To view or add a comment, sign in

More Relevant Posts

Seeed Studio

42,733 followers
4mo Edited
Report this post
🚀 Experience all task models of Ultralytics #YOLO on NVIDIA Robotics Jetson Orin NX device with a single command! 3 steps to easily get started: Install Jetson examples package-> Restart your Jetson device-> Run command 'reComputer run ultralytics-yolo' Key features: 🔍 Detect objects in images, videos, or real-time camera feeds! 📸 Perform object detection, image segmentation, pose estimation, OBB, and classification! 📥 Upload & test your own trained models! Get started now and transform your machine into an AI powerhouse! Check it out on GitHub: https://lnkd.in/gkTp7cei Discover #reComputer J4012 with Jetson Orin NX 16GB device to develop your first Edge AI pipeline: https://lnkd.in/g7J7w7Ts #NVIDIA #JetsonOrin #ultralytics #yolov8 #objectdetection #visionAI #edgeAI #videoanalytics
Like Comment
To view or add a comment, sign in
Seeed Studio

42,733 followers
4mo
Report this post
#ProjectSpotlight 👨💻Build your local RAG system for video retrieval and key points summary on NVIDIA Robotics Jetson Orin! Check out the full project: https://lnkd.in/gGVkPX_Z Project workflow: ✅ Utilized #whisper to process the voice input and convert from voice to text; ✅ Extract images from the video and describe them in the text version; ✅ Embedded RAG under the #llamaindex framework, to prepare for the private vector database. Seamless interaction: ask a question->system retrieves text from the vector database->answer your question based on retrieving the most relevant contents. 📺 Live demo on YouTube: https://lnkd.in/gPhSXERi 👾 Discover more details about reComputer J4012 edge device used in this project: https://lnkd.in/g7J7w7Ts #nvidia #jetson #multimodal #rag #videosearch #edgeai
Like Comment
To view or add a comment, sign in
In-Position Technologies

1,389 followers
6mo
Report this post
The 16th GPU Technology Conference (GTC) witnessed NVIDIA unveil groundbreaking advancements in robotics. Jensen Huang, NVIDIA's co-founder, emphasized the game-changing potential of generative AI, labeling it as the defining technology of our era. Leading the charge is NVIDIA's latest GPU, Blackwell, boasting an impressive 208 billion transistors. This powerhouse is poised to drive the next industrial revolution by powering AI-driven innovations across various industries. But that's not all—digital twins stole the show, offering a cost-effective solution for simulating, testing, and refining large-scale AI deployments before they hit industrial settings. This approach promises to save both time and resources while ensuring the smooth implementation of AI technologies. Teradyne's Universal Robots showcased their AI-powered inspection system, leveraging NVIDIA's Jetson AGX Orin edge AI computer. This system promises to revolutionize robot path planning, making it up to 80 times faster than current systems. Meanwhile, Mobile Industrial Robots A/S unveiled a pallet jack AMR equipped with 3D vision technology, promising unparalleled precision in dynamic environments. The future of robotics is here, and it's more exciting than ever! For deeper insights, dive into the captivating article shared by Robotiq in the comments below. #GTC #AI #Robotics #Innovation

1 Comment
Like Comment
To view or add a comment, sign in
Elaine Wu

Building Edge AI @SeeedStudio, NVIDIA Robotics Elite OEM Partner | Looking to deliver solutions with partners!
4mo
Report this post
#computervision made easy! 🚀 To quickly determine which YOLO model best suits your application dataset in the field, try the Jetson-example. https://lnkd.in/gze_xs86 Firstly >> pip install jetson-examples After reboot >> reComputer run ultralytics-yolo We built this tool and made it the easiest way to test all #YOLOv8 models as well as your custom models!

Seeed Studio

42,733 followers
4mo Edited

🚀 Experience all task models of Ultralytics #YOLO on NVIDIA Robotics Jetson Orin NX device with a single command! 3 steps to easily get started: Install Jetson examples package-> Restart your Jetson device-> Run command 'reComputer run ultralytics-yolo' Key features: 🔍 Detect objects in images, videos, or real-time camera feeds! 📸 Perform object detection, image segmentation, pose estimation, OBB, and classification! 📥 Upload & test your own trained models! Get started now and transform your machine into an AI powerhouse! Check it out on GitHub: https://lnkd.in/gkTp7cei Discover #reComputer J4012 with Jetson Orin NX 16GB device to develop your first Edge AI pipeline: https://lnkd.in/g7J7w7Ts #NVIDIA #JetsonOrin #ultralytics #yolov8 #objectdetection #visionAI #edgeAI #videoanalytics
Like Comment
To view or add a comment, sign in
Liveplex

1,387 followers
6mo
Report this post
At #nvdia GTC, Jensen Huang announced Project GR00T, a general-purpose foundation model for humanoid robots. As part of the initiative, the company also unveiled a new computer, Jetson Thor, for humanoid robots based on the NVIDIA Thor system-on-a-chip (SoC). It also announced significant upgrades to the NVIDIA Isaac™ robotics platform, including generative AI foundation models and tools for simulation and AI workflow infrastructure. Robots powered by GR00T, which stands for Generalist Robot 00 Technology, will be designed to understand natural language and emulate movements by observing human actions — quickly learning coordination, dexterity, and other skills to navigate, adapt, and interact with the real world. Jetson Thor was created as a new computing platform capable of performing complex tasks and interacting safely and naturally with people and machines. It has a modular architecture optimized for performance, power and size. The SoC includes a next-generation GPU based on the NVIDIA Blackwell architecture with a transformer engine delivering 800 teraflops of 8-bit floating point AI performance to run multimodal generative AI models like GR00T.
Like Comment
To view or add a comment, sign in
Ashwin Ned

Applied Science at Fraunhofer IML. Robotics, Deep Learning & 3D Perception
7mo Edited
Report this post
It was incredibly exciting to watch the NVIDIA GTC keynote live today! Here are some takeaways I found interesting for the robotics field. 1. Power & Processing: The new AGX brings low-power SoCs with high-speed sensor processing, allowing robots to run complex AI models like LLMs directly on the hardware. 2. Robotic Learning: Omniverse with OVX creates a 'Gym' for robots to train in simulated environments, accelerating real-world learning and enabling zero-shot transfer from simulation. 3. Smart Warehouses: Imagine warehouses where everything - machines, objects, even the building itself - acts as an autonomous system, managed by an AI "air traffic controller." Nvidia Metropolis is making this a reality. 4. Jetson Orin: This powerful robotics computer along with it's scalable software stack and AI microservices delivers exceptional performance for on-device processing for the next generation of energy efficient autonomous machines. 5. Nvidia Isaac Gets Smarter: Isaac Perceptor SDK offers built-in perception capabilities like visual odometry, 3D reconstruction, and depth perception. Isaac Manipulator leverages geometric awareness, motion planning with CUDA, and 3D object pose estimation for optimal grasping. 6. Generalizing Humanoid Robotics: Nvidia introduced Project GROOT, a foundation model equipped with new Jetson THOR chips that is capable of learning from multimodal data and outputs actions. The model is trained on Issac RL gym using the Issac robotics stack and offers zero-shot transfer capabilities in humanoid robots such as that from Agility Robotics. This is just a glimpse into the exciting future of robotics! Can't wait to see what the rest of 2024 will bring to robotics. #Nvidia #GTC #AI #Robotics
Like Comment
To view or add a comment, sign in
Seeed Studio

42,733 followers
5mo Edited
Report this post
Accelerate local LLM deployment with Llama 3 on Jetson Orin! With just one simple command, you can easily deploy Llama 3 with 8 billion parameters directly onto #reComputer edge devices powered by NVIDIA Robotics Jetson Orin. Use the following command to get started: >>reComputer run llama3 🏃♂️ Exercise impressive performance with up to 40 tokens/second on Jetson AGX Orin and 15 tokens/second on Jetson Orin Nano. This setup is ideal for creating local Retrieval Augmented Generation (RAG) systems, enhancing interactive robotics, and developing other cutting-edge Generative AI applications. 🔍 Discover more edge AI possibilities through our Jetson examples library: https://lnkd.in/gPcgtSuq 📖 Learn more about NVIDIA Accelerates Inference on Meta Llama 3: https://lnkd.in/gQJPaduh 🛒 Looking to get started? Find the reComputer Orin Nano edge device and AGX Orin developer kit here: https://lnkd.in/emYBcqUQ https://lnkd.in/gvyk5QAf #nvidia #jetson #metaai #llama3 #generativeai #llm
3 Comments
Like Comment
To view or add a comment, sign in
leetop

10 followers
7mo
Report this post
🚀 Elevate Your Innovation with the A206 Carrier Board for NVIDIA Jetson Nano! 🤖 Attention innovators, creators, and tech enthusiasts! Unleash the full potential of your NVIDIA Jetson Nano with the ultimate expansion: the A206 Carrier Board. Designed for versatility and superior performance, this carrier board is your key to unlocking endless possibilities in your projects. 🌟 Why Choose the A206 Carrier Board? Versatile I/O Options: Equipped with multiple input/output options to enhance connectivity and performance. Fully Customizable: Seamlessly integrates with your existing setups, making it perfect for a wide range of applications. Empower Your Applications: From robotics to AI research and educational projects, the A206 brings your ideas to life. Whether you're building the next generation of smart robots, diving deep into AI research, or developing educational tools, the A206 Carrier Board is designed to elevate your projects to new heights. 🛠️💡 🔗 Ready to start innovating? Discover more about the A206 Carrier Board and how it can transform your NVIDIA Jetson Nano projects. Begin your journey to innovation today! #A206CarrierBoard #NVIDIAJetsonNano #TechInnovation #Robotics #AIResearch #EducationalProjects
Like Comment
To view or add a comment, sign in
Seeed Studio

42,733 followers
6mo Edited
Report this post
#GTC2024 😎 moments review: Join Roboflow and explore how to seamlessly deploy YOLO-World zero-shot object detection model on #reComputer Industrial edge device powered by NVIDIA Robotics Jetson Orin Nano! The whole pipeline transcends traditional boundaries with its open vocabulary capability, allowing it to recognize objects beyond predefined categories—making it more efficient and adaptable for real-world applications. Experience the flexibility of dynamically adjusting detection vocabularies to meet diverse needs without compromising performance. 👉 To get hands on the YOLO-World approach on Jetson Orin Nano 8GB, you may want to check out our reComputer Industrial J3011: https://lnkd.in/gcfq6g8D 🏂 YOLO-World GitHub repo: https://lnkd.in/gvASzQUX #nvidia #jetson #computervision #edgeai #zeroshot #objectdetection #yoloworld
Like Comment
To view or add a comment, sign in
Seeed Studio

42,733 followers
6mo Edited
Report this post
#ProjectSpotlight Discover the power of NVIDIA Robotics Jetson Orin in transforming spoken language into written text in real-time! Our latest wiki shows guidance of implementing the Whisper AI model on the #reComputer Industrial Jetson Orin Nano edge device. Dive into the wiki details here: https://lnkd.in/g7JsEwbi This wiki demonstrates how NVIDIA Jetson Orin brings the #Whisper model to life with the great support of #TensorRT acceleration, providing an efficient and accurate real-time speech-to-text conversion. While Whisper may experience some delays at the beginning of the task, it still provides about the same accuracy and efficiency as RIVA based on its relatively simple deployment process. Discover more possibilities with #reComputer Industrial Jetson Orin Nano edge device: https://lnkd.in/gczuEfR6 and #reSpeaker mic: https://lnkd.in/g2EipSYV #nvida #jetson #wiki #whisper #respeaker #ASR #STT #speechtotext
Like Comment
To view or add a comment, sign in

Seeed Studio

42,733 followers

View Profile Follow

More from this author

Explore topics

翻译：