Groq’s Stunning Emergence Time flies, and many of you may remember that last year, in January and February, an AI company with a rather peculiar name, Groq, made headlines with its self-developed LPU (Language Processing Unit) chip system, showcasing extreme low-latency and high-performance inference on the Llama2 70B model (with actual performance exceeding GPU clusters … Read more
As artificial intelligence (AI) models become increasingly complex and widespread, the industry continues to seek the most effective hardware to meet the evolving demands of AI inference. While GPUs, TPUs, and CPUs have traditionally handled various AI workloads, FPGAs (Field-Programmable Gate Arrays)— especially with the support of high-performance architectures like the Achronix Speedster7t FPGA — … Read more
Recently, an investor asked Loongson Technology on an interactive platform whether the 9A1000 has successfully taped out.On September 15, Loongson Technology responded that the development of its first GPGPU chip, the 9A1000, is basically complete, and tape-out will be delivered within the third quarter. The success of the tape-out will depend on the test results … Read more
Have you ever thought about building your own personal server using a computer the size of a credit card? This is not a scene from a science fiction movie, but a reality! With the Raspberry Pi, you can easily create a private cloud storage, media center, AI inference platform, or an automated office assistant. This … Read more
1. What is an NPU? NPU (Neural Processing Unit) is a processor specifically designed for inference and training of artificial neural networks, featuring high parallelism, low power consumption, and low latency. It is a key hardware component in scenarios such as edge AI, autonomous driving, smart cameras, and voice recognition. 2. Comparison of NPU with … Read more
Loongson Technology has announced that its first graphics card, the 9A1000, has completed development and is expected to enter the tape-out stage in the third quarter of this year. Unlike the gaming graphics cards we are familiar with, the 9A1000 is a General-Purpose Graphics Processing Unit (GPGPU), which is not focused on rendering images but … Read more
ASIC is a “custom chip”—unlike GPUs that can handle various tasks, it is optimized for specific functions (such as AI inference and firewall protection), achieving 40% higher efficiency and 3-5 times lower power consumption compared to general-purpose chips. For example, when you chat with AI, the underlying “inference computation” relies on ASICs; 5G base stations … Read more
With the expansion of the artificial intelligence inference market, global tech giants are reducing their reliance on GPUs (Graphics Processing Units). They are seeking next-generation alternatives to the expensive and heat-generating GPUs. The Neural Processing Unit (NPU) is an emerging alternative, specifically designed for AI inference. This architecture, tailored for AI computation, achieves high efficiency … Read more
The scenarios of algorithm solidification and large-scale deployment of inference are eroding the market originally belonging to GPUs, with ASIC chips becoming the new favorite of tech giants due to their extreme energy efficiency and low-cost advantages. By 2025, global tech giants will face a common challenge: how to efficiently handle the explosive growth of … Read more
With the expansion of the artificial intelligence inference market, major global technology companies are reducing their reliance on GPUs (Graphics Processing Units). They are seeking next-generation alternatives to the expensive and heat-generating GPUs. The Neural Processing Unit (NPU) is an emerging alternative, specifically designed for AI inference. This architecture, tailored for AI computation, achieves high … Read more