Friday, 31 January 2025
28.1 C
Singapore
27 C
Thailand
21.3 C
Indonesia
27.1 C
Philippines

Microsoft supports new startups as it launches an AI processor that skips GPUs and expensive memory

Discover how d-Matrix's new AI processor, Corsair, backed by Microsoft, redefines AI inference with a GPU-free and cost-effective solution.

D-Matrix Inc., a hardware startup based in Santa Clara, California, has unveiled its first AI processor, Corsair, which aims to redefine AI inference. This innovative product does away with traditional GPUs and costly high-bandwidth memory (HBM), offering a more efficient and cost-effective solution.

Supported by Microsoft and embracing cutting-edge advancements, Corsair is available to early-access customers and is expected to reach broader availability by the second quarter of 2025.

What makes Corsair stand out?

Corsair is purpose-built to tackle demanding AI inference tasks, particularly those involving generative AI models. The processor achieves remarkable speeds, handling 60,000 tokens per second at 1 millisecond per token when running models like Llama3 8B on a single server.

For larger-scale applications, such as Llama3 70B, Corsair achieves 30,000 tokens per second at 2 milliseconds per token on a single rack. This performance significantly lowers energy consumption and operational costs compared to traditional GPU-based systems.

Built on d-Matrixโ€™s Nighthawk and Jayhawk II tiles, Corsair utilises a 6nm manufacturing process. Each Nighthawk tile has four neural cores and a RISC-V CPU, optimised for large-model inference. It incorporates digital in-memory computation (DIMC) and supports versatile datatype processing, including block floating point (BFP).

Corsairโ€™s chiplet packaging integrates memory and computation to boost efficiency. It adheres to the PCIe Gen5 full-height, full-length card form factor, making it compatible with DMX Bridge cards for scalable performance. The card is powered by 2400 TFLOPs of 8-bit peak computing power, 2GB of integrated performance memory, and 256GB of off-chip memory capacity.

A partnership with Nvidiaโ€™s key ally

Micron Technology, a significant partner of Nvidia, is collaborating with d-Matrix to support Corsairโ€™s development and growth.

Although initially slated for release in late 2023, d-Matrix restructured its architecture to align with the growing demand for generative AI. This shift allowed Corsair to incorporate advanced features tailored for transformer models, agentic AI, and emerging applications like interactive video generation.

โ€œOur vision for d-Matrix was to address the massive computing challenges of generative AI and transformers,โ€ said Sid Sheth, co-founder and CEO of d-Matrix. โ€œCorsair is a groundbreaking platform, delivering blazing-fast token generation for interactive AI applications, making generative AI commercially viable.โ€

Corsairโ€™s focus on cost efficiency, scalability, and high-performance positions it as a promising solution in the rapidly evolving AI landscape.

Hot this week

X rolls out vertical video feed to global iOS users

X expands its vertical video feed globally for iOS users, aiming to compete with TikTok and increase ad revenue through engaging video content.

iOS 18.3 arrives with AI notification summary tweaks

iOS 18.3 changes AI notification summaries, including new settings and improved features for Apple devices.

Samsung could add optical blood glucose monitoring to the Galaxy Watch series

Samsung may add an optical blood glucose sensor to its Galaxy Watch, offering non-invasive monitoring and challenging Apple in wearable health tech.

Meta to invest US$65 billion in AI infrastructure in 2025

Meta will invest US$65 billion in AI by 2025, building massive data centres and expanding AI teams to lead cutting-edge innovation.

Sony extends InZone monitor warranty to three years with OLED burn-in coverage

Sony now offers a three-year warranty with OLED burn-in coverage for its InZone M10S monitors, addressing long-term concerns for gamers.

Apple CEO praises DeepSeekโ€™s AI despite controversy

Apple CEO Tim Cook praises DeepSeekโ€™s AI despite OpenAIโ€™s allegations, while Apple Intelligence faces a slow start and AI news summaries spark controversy.

Nvidia’s DLSS 4 brings enhanced image quality and efficiency

Nvidiaโ€™s latest GPU driver update brings DLSS 4 to unsupported games, improves video upscaling, and introduces Smooth Motion for RTX 50-series owners.

Appleโ€™s revenue rises despite an 11% drop in China sales

Appleโ€™s Q1 2025 revenue rose 4% to US$124.3B, despite an 11% decline in China iPhone sales. Strong growth in services and Mac sales helped offset losses.

Pentagon moves to block DeepSeek after staff access Chinese servers

The Pentagon is blocking DeepSeek after employees unknowingly connected work computers to Chinese servers, raising national security concerns.

Related Articles