https://yzma.ai/devices/ Recent content in Devices on yzma Hugo en-us https://yzma.ai/devices/arduino-uno-q/ Mon, 01 Jan 0001 00:00:00 +0000 https://yzma.ai/devices/arduino-uno-q/ <img src="https://yzma.ai/images/arduino-logo.png" alt="Arduino logo" class="platform-logo"> <p>The Arduino UNO Q is a unique board with two functions. It has both a Qualcomm QRB2210 arm64 processor running a full Debian based Linux, as well as a STM32U585 microcontroller.</p> <p>yzma runs on the Linux side of the board and uses the CPU for inference. With a small text model, the board can process about 32 tokens a second.</p> <figure class="device-photo"> <img src="https://yzma.ai/images/devices/arduino-uno-q.webp" alt="Arduino UNO Q board"> <figcaption>Image by <a href="https://github.com/arduino/docs-content">Arduino</a>, <a href="https://creativecommons.org/licenses/by-sa/4.0/">CC BY-SA 4.0</a></figcaption> </figure> <h2 id="links">Links</h2> <ul> <li><a href="https://docs.arduino.cc/hardware/uno-q/">Arduino UNO Q website</a></li> <li><a href="https://yzma.ai/getting-started/install/arduino-uno-q/">Getting started on the Arduino UNO Q</a></li> <li><a href="https://yzma.ai/docs/reference/benchmarks/#text-generation">Text generation benchmarks</a></li> <li><a href="https://yzma.ai/docs/reference/benchmarks/#multimodal">Multimodal benchmarks</a></li> </ul> https://yzma.ai/devices/jetson-orin-nano/ Mon, 01 Jan 0001 00:00:00 +0000 https://yzma.ai/devices/jetson-orin-nano/ <img src="https://yzma.ai/images/NVIDIA-logo.png" alt="NVIDIA logo" class="platform-logo"> <p>The NVIDIA Jetson Orin Nano is a small computer for AI at the edge. It has an arm64 processor and an NVIDIA GPU on the same board.</p> <p>yzma uses the GPU with CUDA or Vulkan. With a small text model and CUDA, the board can process about 190 tokens a second.</p> <figure class="device-photo"> <img src="https://yzma.ai/images/devices/jetson-orin-nano.webp" alt="NVIDIA Jetson Orin Nano Developer Kit"> <figcaption>Image by <a href="https://www.seeedstudio.com/NVIDIAr-Jetson-Orintm-Nano-Developer-Kit-p-5617.html">Seeed Studio</a></figcaption> </figure> <h2 id="links">Links</h2> <ul> <li><a href="https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-orin/nano-super-developer-kit/">NVIDIA Jetson Orin Nano website</a></li> <li><a href="https://yzma.ai/getting-started/install/jetson-orin/">Getting started on the Jetson Orin</a></li> <li><a href="https://yzma.ai/docs/reference/benchmarks/#text-generation">Text generation benchmarks</a></li> <li><a href="https://yzma.ai/docs/reference/benchmarks/#multimodal">Multimodal benchmarks</a></li> </ul> https://yzma.ai/devices/raspberry-pi/ Mon, 01 Jan 0001 00:00:00 +0000 https://yzma.ai/devices/raspberry-pi/ <img src="https://yzma.ai/images/raspberry-pi-os-logo.png" alt="Raspberry Pi logo" class="platform-logo"> <p>The Raspberry Pi is a popular single board computer. yzma runs on a Raspberry Pi 4 and on a Raspberry Pi 5 with the 64 bit version of the Raspberry Pi OS.</p> <p>yzma uses the CPU. With a small text model, a Raspberry Pi 4 can process about 35 tokens a second.</p> <figure class="device-photo"> <img src="https://yzma.ai/images/devices/raspberry-pi-5.webp" alt="Raspberry Pi 5 board"> <figcaption>Image by <a href="https://github.com/raspberrypi/documentation">Raspberry Pi Ltd</a>, <a href="https://creativecommons.org/licenses/by-sa/4.0/">CC BY-SA 4.0</a></figcaption> </figure> <h2 id="links">Links</h2> <ul> <li><a href="https://www.raspberrypi.com/">Raspberry Pi website</a></li> <li><a href="https://yzma.ai/getting-started/install/raspberry-pi/">Getting started on the Raspberry Pi</a></li> <li><a href="https://yzma.ai/docs/reference/benchmarks/#text-generation">Text generation benchmarks</a></li> <li><a href="https://yzma.ai/docs/reference/benchmarks/#multimodal">Multimodal benchmarks</a></li> </ul>