Well, after entering the discrete GPU market, Moore Threads has also taken its chance in the APU segment, showcasing its high-end SoC for laptops.
Moore Threads' New Laptop Chip Offers Impressive Edge AI Performance, Rivaling Current-Gen Lineups
Moore Threads have been a popular name in our coverage, and in many of them, they have been known for coming up with rather interesting solutions that turn out to be pretty interesting. Like, one example of this is how the Chinese GPU manufacturer was one of the first to showcase a PCIe 5.0 GPU, which eventually turned out to be slower than NVIDIA's GTX 1050 Ti. And now, it seems the manufacturer has entered the laptop SoC segment with its in-house 'Yangtze' architecture, which combines a native GPU, software capabilities, and CPU microarchitectures.
Diving a bit into the technicals, the Yangtze chip offers a 12-core configuration under the 'ARMv8' architecture, along with an iGPU featuring the company's in-house MUSA architecture. More specifically, the chip features a base frequency of 2.65 GHz and is integrated into the MTT AIBOOK, which is, yet again, an OEM offering from Moore Threads. In terms of the laptop specifications, the AIBOOK offers a 32 GB DDR5-7500 memory, along with a 1 TB SSD capacity, with an OLED panel that is rated at 120 Hz refresh rate.
The laptop itself is a solid offering, dedicated to mid-to-high-end workloads, but the more interesting aspect is indeed the self-built chip onboard. We did talk about this chip in the past as well, but back then, it was limited to an 8-core configuration, so somehow, the manufacturer has decided to bump up the specifications. Another interesting aspect is the onboard NPU, which squeezes out 50 TOPS (INT8), in a multi-core neural processor package, designed for speech/image recognition. Here are the key highlights of the chip itself:
- 12-core CPU
- 2.65 GHz Boost Clock
- Integrated GPU Based on High-End IP
- 50 TOPS NPU
- H.265/H.264/AV1 Support
- Aims AI PC Segment
The AIBOOK platform is Moore Threads' effort to capitalize on the AI PC hype in China, especially since deploying models on local hardware has gained immense traction in the region, driven by the nation's pursuit of open-source LLMs. The chip does look great on paper, but the real question is how it performs in real-world benchmarks, given that we cannot be certain of its performance based solely on paper specifications.
In terms of pricing, the MTT AIBOOK is listed at around $1,440 on China's JD.com, which is an expensive price tag for such a device, but it reflects the manufacturer's first entry into the APU segment.
News Source: Videocardz
About the author: Muhammad Zuhair is a hardware and technology reporter for Wccftech, specializing in the semiconductor industry and the complex interplay between technology, manufacturing, and geopolitics. His coverage focuses on the corporate strategies and technological roadmaps of industry giants like TSMC, NVIDIA, Samsung, and Intel. Zuhair's expertise lies in deconstructing complex topics such as fabrication nodes (e.g., 2nm process), the economic impact of policies like the CHIPS Act, and the strategic development of AI infrastructure from NVIDIA, AMD and Intel.
Follow Wccftech on Google to get more of our news coverage in your feeds.
