Lambda secures a cloud agreement with Hudson River Trading to provide access to NVIDIA chips.

      The GPU-cloud startup preparing for an IPO has added one of the largest quantitative trading firms in the United States to its customer base, which already includes Microsoft and NVIDIA. Lambda has reached a cloud infrastructure agreement with Hudson River Trading (HRT) to provide access to NVIDIA chips, as reported by Reuters on Wednesday.

      This deal expands Lambda's clientele in the high-frequency trading sector as it gears up for an IPO anticipated in the first half of 2026. HRT now joins the ranks of Lambda's customers that feature Microsoft and NVIDIA. In November 2025, Lambda revealed a multibillion-dollar partnership with Microsoft involving tens of thousands of NVIDIA GPUs, including the GB300 NVL72 systems. Meanwhile, NVIDIA has its own $1.5 billion, four-year agreement to lease roughly 18,000 of its GPUs back from Lambda, making NVIDIA its largest individual customer.

      Adding HRT to this existing customer base enhances Lambda's presence in the latency-sensitive area of the GPU market, which typically commands higher prices than model-training workloads. HRT has shown a longstanding interest in premium NVIDIA capacity, having operated an NVIDIA-driven AI factory for its algorithmic trading research that utilizes Blackwell architecture and Spectrum-X networking. Moreover, it has been engaged in a Google Cloud partnership for trading simulation workloads since 2024.

      By incorporating Lambda as a third computing provider, HRT aims to maintain a multi-vendor procurement approach, mitigating the risks of capacity constraints from relying on a single hyperscaler amid the ongoing GPU allocation challenges. The rationale for HRT's procurement strategy aligns with its scale; it reported record quarterly trading revenue of $6.4 billion in the first quarter of 2026, following a full-year trading revenue of $12.3 billion in 2025. Consequently, a multi-million-dollar GPU procurement commitment represents a relatively minor expense compared to its overall revenue, with the critical factor being the availability of capacity when needed by the research team.

      Lambda has specifically targeted AWS, Azure, and Google Cloud regarding time-to-capacity rather than pricing, focusing on shorter procurement cycles and dedicated allocations as key differentiators.

      This win for Lambda arrives at a particularly advantageous time in relation to its IPO narrative, as the company has enlisted Morgan Stanley, JP Morgan, and Citi for an offering scheduled for the first half of 2026, according to various secondary-market reports. Featuring the HRT logo alongside Microsoft and NVIDIA on its customer slide enhances Lambda's appeal by broadening its narrative beyond its historical focus on model-lab and hyperscaler-extension verticals.

      The impact of this deal on Lambda's bookings ahead of the IPO will hinge on the contract structure, which Reuters characterized as a multi-year supply agreement without revealing the total contract value.

      This agreement exists within the broader context of GPU allocation issues. Recent developments, such as Google's $25 billion TPU-cloud joint venture with Blackstone and Tenstorrent's discussions to acquire Intel and Qualcomm, reflect the same structural shortages in the market. The current capacity-allocation challenges have led sophisticated commercial buyers like HRT to diversify their sources, opting for multiple compute providers instead of relying on just one.

      The complexities exacerbated by export controls, including the licensing path for H200 sales under the Trump-Xi era, are contributing to the ongoing volatility in the supply landscape as the year progresses. Lambda has not disclosed the contract value, the specific generations of NVIDIA chips involved (Blackwell, GB300, or earlier H-series), the geographical distribution of the provided capacity, or the contract duration beyond its multi-year term. HRT has not issued a public statement regarding the announcement.

      The next significant indicator will be Lambda's S-1 filing, anticipated prior to the intended first-half IPO, which will provide insights into customer concentration, allowing a clearer assessment of how significant the HRT contract is compared to those with Microsoft, NVIDIA, and previous model-lab agreements.