Raspberry Pi gets a serious competitor as Particle unveils its 5G-capable SBC
Particle has launched the Tachyon, a $149 single-board computer featuring 5G, AI capabilities, and advanced specs, aiming to democratize technology for engineers and innovators, available on Kickstarter.
Read original articleParticle has introduced the Tachyon, a new single-board computer (SBC) that competes directly with the Raspberry Pi by offering advanced features such as 5G connectivity and AI capabilities. Priced at $149, the Tachyon is designed to be accessible for engineers and innovators, aiming to democratize cutting-edge technology. The board is compact, roughly the size of a credit card, and includes an octa-core Qualcomm Kryo CPU, 4GB of RAM, and 64GB of flash storage. It also features an Adreno 643 GPU, 12 TOPS of AI processing power, and over 20 camera sensors, along with a USB-C 3.1 PD port and a PCIe Gen 3 slot.
Zach Supalla, CEO of Particle, emphasized the importance of making advanced technology available to a broader audience, stating that current access to 5G and AI accelerators is limited to a few large companies. The Tachyon aims to inspire innovation by providing essential tools for developing connected devices. Interested individuals can support the project through Kickstarter, where Particle has a history of successful campaigns. The Tachyon's introduction highlights the growing competition in the SBC market, showcasing the potential for new functionalities that go beyond what traditional boards like the Raspberry Pi offer.
Related
Supercomputer-on-a-chip goes live: single PCIe card packs more than 6k cores
InspireSemi introduces Thunderbird I Accelerated Computing chip with 1,536 custom RISC-V CPU cores for scientific and data processing. Energy-efficient, scalable to 360,000 cores, suitable for high-performance computing tasks. CEO praises team's work.
What Would You Do with a 16.8M Core Graph Processing Beast?
TSMC collaborates with Intel, MIT, and AWS to develop HIVE, a graph processing unit named PIUMA with 16.8 million cores. The advanced chip aims to efficiently process neural networks, featuring a custom RISC-based architecture and photonics interconnect for scalability.
Tenstorrent Wormhole Developer Kits Launched
Tenstorrent launches "Wormhole" PCIe developer kits with n150s and n300s cards, Tensix cores, GDDR6 memory, local SRAM caches, and high-speed Ethernet. Priced at $18K, offering cost-saving opportunities for AI developers.
Radxa X4 low-cost, credit card-sized Intel N100 SBC goes for $60 and up
The Radxa X4 is a compact Intel Processor N100 SBC with 4GB/8GB RAM, dual HDMI, USB, Ethernet, WiFi, GPIO, RP2040 microcontroller, M.2 SSD support, Windows/Linux compatibility. Lacks camera/display interfaces. Compatible with Raspberry Pi GPIO. Out of stock temporarily.
The Luckfox Pico Mini B – Linux in a Thumbnail
Luckfox released the Pico Mini B, a compact Linux-capable development board with Cortex A7 core, NPU, 64MB RAM, USB-C, CSI connector, and 17 GPIO pins for $8.99. It supports Python, C, and GPIO control, appealing to embedded system enthusiasts and developers.
If running some AI model, like llama3.1, it requires NVidia/AMD GPU and more than 16 GB RAM. I have run it at i7 laptop, that has 32 GB RAM and NVidia GPU, using Ubuntu 24.04, with commands:
curl -fsSL https://ollama.com/install.sh | sh
ollama run llama3.1
Without supported GPU, it will run in slower CPU mode.
Running at M1 Air 16 GB RAM, at Asahi Ubuntu arm64, it runs at CPU mode. I have not yet tried at macOS.
There is some SBC that has 16 GB or more RAM, like OrangePi, but I have not yet tried can they run llama3.1 .
What kind of AI can run with 4 GB RAM ?
> The Tachyon also sports an octa-core Qualcomm Kryo CPU, 4GB RAM, and 64GB of built-in flash storage.
They want to market it for "AI" use cases, but only include 4GB of RAM?
https://www.kickstarter.com/projects/1598272670/chip-the-wor...
Wow.
Related
Supercomputer-on-a-chip goes live: single PCIe card packs more than 6k cores
InspireSemi introduces Thunderbird I Accelerated Computing chip with 1,536 custom RISC-V CPU cores for scientific and data processing. Energy-efficient, scalable to 360,000 cores, suitable for high-performance computing tasks. CEO praises team's work.
What Would You Do with a 16.8M Core Graph Processing Beast?
TSMC collaborates with Intel, MIT, and AWS to develop HIVE, a graph processing unit named PIUMA with 16.8 million cores. The advanced chip aims to efficiently process neural networks, featuring a custom RISC-based architecture and photonics interconnect for scalability.
Tenstorrent Wormhole Developer Kits Launched
Tenstorrent launches "Wormhole" PCIe developer kits with n150s and n300s cards, Tensix cores, GDDR6 memory, local SRAM caches, and high-speed Ethernet. Priced at $18K, offering cost-saving opportunities for AI developers.
Radxa X4 low-cost, credit card-sized Intel N100 SBC goes for $60 and up
The Radxa X4 is a compact Intel Processor N100 SBC with 4GB/8GB RAM, dual HDMI, USB, Ethernet, WiFi, GPIO, RP2040 microcontroller, M.2 SSD support, Windows/Linux compatibility. Lacks camera/display interfaces. Compatible with Raspberry Pi GPIO. Out of stock temporarily.
The Luckfox Pico Mini B – Linux in a Thumbnail
Luckfox released the Pico Mini B, a compact Linux-capable development board with Cortex A7 core, NPU, 64MB RAM, USB-C, CSI connector, and 17 GPIO pins for $8.99. It supports Python, C, and GPIO control, appealing to embedded system enthusiasts and developers.