Senior Software Engineer - Kernels

Nvidia to Invest in Chip Rival d-Matrix as Its Challengers Choose Partnership

d-Matrix announced its Corsair AI data centre platform has been named to Fast Company's 2026 Next Big Things in Tech list. The platform is now shipping to priority customers. Corsair was recognised in the Foundational AI category for its memory-centric design, which brings compute directly into memory to eliminate data-movement bottlenecks. The approach aims to enable faster, more energy-efficient AI at scale. Fast Company's annual list recognises breakthrough technologies with potential to transform industries. This year's honourees span multiple industries, each demonstrating potential for significant impact. d-Matrix develops memory-centric compute chips, software, and rack-scale systems for data centres. The company works independently or alongside GPUs and other compute platforms to address cost, latency, and power limitations of traditional architectures.

d-Matrix's Raptor puts stacked DRAM to work on inference bandwidth. An ISCA paper reports up to 100 TB/s per card, while its throughput comparisons remain modeled and Raptor is still pre-production. By Ryan Merket - Published Oct 2nd, 2026, 7:41pm CT - 4 MIN READ - 6 SOURCES - VIA Aligned News - AI Intelligence Listen to article. 5 min 59 sec d-Matrix's Raptor accelerator puts compute logic face-to-face with stacked DRAM to address the data-movement bottleneck in generative inference. A paper presented at ISCA 2026 reports up to 100 TB/s of bandwidth per card and higher throughput than HBM- and SRAM-based configurations under modeled deployment assumptions. Those are research results, not production benchmarks against shipping accelerators. The figures resurfaced in an October 2nd post on X, but the underlying research was presented at ISCA in Raleigh, North Carolina, during the conference's June 27th through July 1st run. d-Matrix CTO Sudeep Bhoja also presented the technology at Hot Chips in August. The work describes early Raptor silicon; d-Matrix says the commercial product is expected in the fourth quarter of 2027. For co-founder and CEO Sid Sheth, Raptor extends a bet made before generative AI turned inference into a mainstream infrastructure priority. Sheth wrote that he and co-founder Sudeep Bhoja left Inphi in 2019 after seeing cloud customers use its data-center interconnect products in distributed AI systems. They concluded that running trained models at scale would become a major computing market, even while investors were still focused on training. A memory bet from the start. Sheth's background is in the movement of data through networks and data centers. Before founding d-Matrix, he ran Inphi's networking business; Bhoja, now d-Matrix's CTO, also brought experience in networking and interconnects. They based their inference thesis on data-center AI workloads and the infrastructure built to serve them, before generative AI became a mainstream infrastructure priority. The bet has taken time to reach product silicon. In a 2024 account of d-Matrix's early years, Sheth recounted an early architecture pivot and the difficulty of keeping hardware and software development aligned. Raptor carries that original memory-focused premise into its physical design: d-Matrix says it places a TSMC 4-nanometer logic die face-to-face with a custom DRAM die. That design targets the repeated movement of model weights and key-value cache data during autoregressive decoding. The Raptor paper reports stream-blocking to map KV-cache streams across configurable DRAM channels, alongside techniques for power, thermal management and error correction. Its workloads include Llama 3.1 70B, DeepSeek-V3, Kimi K2, GPT-OSS, Whisper and Canary. The paper reports 4.71 times the throughput of an HBM configuration and 2.44 times that of an SRAM configuration across those workloads. The comparison is modeled, so it does not establish that a Raptor card outperforms a specific Nvidia or AMD product in a customer deployment. The up-to-100-TB/s bandwidth figure is also only part of the system tradeoff: d-Matrix's reported Raptor configuration has 32GB of memory, far less capacity than some HBM-based accelerators. From chip architecture to a rack strategy. Raptor's commercial case now depends on more than the memory stack. In September, d-Matrix announced that it would integrate its next-generation accelerators into Nvidia's MGX rack architecture using NVLink Fusion. The partnership announcement positions Raptor alongside Nvidia GPUs in a system that could assign compute-intensive prefill work to GPUs and latency-sensitive decoding to d-Matrix accelerators. That gives d-Matrix a route into an established rack ecosystem, while making integration, software and deployment economics part of the test alongside chip performance. D-Matrix says Raptor is expected to tape out before the end of 2026, with initial availability in Nvidia MGX racks expected in the fourth quarter of 2027. Those milestones leave manufacturing yield, production volume, pricing and customer performance to be demonstrated. The design's memory capacity also raises a question: how workloads divide between Raptor's high-bandwidth local memory and other memory in a rack. In November 2025, d-Matrix said it closed a $275 million Series C at a $2 billion valuation, bringing its disclosed funding total to $450 million. The round was co-led by BullhoundCapital, Triatomic Capital and Temasek, with participation from Qatar Investment Authority, EDBI, Microsoft's M12, Nautilus Venture Partners, Industry Ventures and Mirae Asset, according to d-Matrix's funding announcement. The money backs the thesis Sheth has pursued since 2019: inference hardware should be designed around the cost of moving data, not simply the volume of arithmetic. Raptor's early silicon and research comparisons give that thesis a more concrete form. Its commercial test will be whether the bandwidth advantage survives the constraints that matter in a data center: enough capacity for useful workloads, reliable production, software compatibility and a system customers can deploy. RuntimeWire, Inc. get virtually zero search traffic. Please consider sharing this story. Signal check Noise Signal Where does this story land? Drag the dial. 1 reader has rated The startup MoonshotAI: Kimi K2 0711. moonshotai Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. View on RuntimeWire $0.570 in · $2.30 out /M 131K context Cutoff 2024-12-31 DeepSeek: DeepSeek V3. deepseek-ai DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. View on RuntimeWire $0.257 in · $1.03 out /M 164K context Cutoff 2024-07-31 OpenRouter: Fusion. OpenRouter Fusion turns your prompt into a small multi-model deliberation. Llama 3.1 70B. meta-llama DeepSeek-V3. deepseek-ai Hugging Face model repository deepseek-ai/DeepSeek-V3, listed under the text-generation task. This listing describes a public repository, not an inference endpoint. Modalities. Stay on the wire. Breaking startup & AI news, funding scoops, and launches as they happen. Follow @runtimewire on Threads and LinkedIn. Free forever - no credit card, no nagging.