Lambda Labs

Lambda Labs (operating as Lambda, Inc.) is an American AI infrastructure company that provides GPU cloud computing, on-premises GPU hardware, and deep learning software for artificial intelligence research and development. Headquartered in San Francisco, California, Lambda operates what it calls a "Superintelligence Cloud," offering on-demand GPU instances, large-scale multi-node clusters, and an inference API to academic institutions, AI startups, and large enterprises.^[1] The company was originally founded in 2012 as a computer vision and facial recognition software provider before pivoting to GPU cloud infrastructure around 2017 to 2018.^[1] Lambda said in March 2025 that it had more than 150,000 cloud users,^[8] and Sacra Research estimated its annualized revenue at approximately $760 million by December 2025, up roughly 79 percent year over year.^[14] As of November 2026 Lambda remained private: in October 2026 The Wall Street Journal reported that it was raising up to $4 billion at a $14.5 billion pre-money valuation ahead of an IPO planned for 2027.^[40] Lambda is widely regarded as one of the largest pure-play GPU neocloud providers, often grouped with CoreWeave, Crusoe, and a handful of other specialists that own and operate their own NVIDIA accelerator fleets rather than reselling hyperscaler capacity. The company states its mission is "to make compute as ubiquitous as electricity and give everyone the power of superintelligence."^[17]

History

When was Lambda founded and by whom?

Lambda was founded in 2012 by twin brothers Stephen Balaban and Michael Balaban.^[1] Stephen Balaban studied computer science and economics at the University of Michigan and had previously been an early engineer at Perceptio, a startup focused on on-device neural networks for facial recognition that Apple acquired in 2015 (after Stephen had already left to start Lambda).^[1] Michael Balaban completed a double major at the University of Michigan in discrete mathematics and computer science, graduated a year after his brother, and joined Nextdoor as a software engineer on the infrastructure team before coming on full time at Lambda in March 2015.^[16] Michael Balaban became the company's CTO while Stephen served as CEO, a division of roles the brothers maintained into the mid-2020s.^[1]

The company's initial product focus was facial recognition software. An early trigger for their first commercial offering came in June 2012 when Facebook acquired Face.com and shut down its popular facial recognition API, stranding roughly 45,000 developers who had built applications on top of it.^[1] Lambda moved quickly to launch the Lambda Face API as an alternative, attracting more than 1,000 active developers within a year and processing over 5 million API calls per month.^[1] During this period the company also offered a facial recognition API for Google Glass developers, which drew controversy.^[14]

In 2015 the brothers launched a consumer app that applied the neural artistic style technique to selfies; after a Reddit post it at one point drew 30,000 sign-ups a day, and its Amazon Web Services bill grew to about $40,000 a month.^[16] Lambda's own history also describes early facial recognition research "supported by a camera in a hat."^[8] Although the early business never scaled to the level of consumer giants, it gave the founders deep familiarity with GPU-accelerated deep learning workloads, which would later become central to Lambda's pivot.^[16]

Pivot to GPU hardware and cloud (2017 to 2019)

As deep learning workloads became more GPU-intensive, the Balaban brothers found that running their own machine learning models on AWS was prohibitively expensive and operationally cumbersome.^[16] They built an internal GPU cluster to support their own workloads, then recognized that other AI teams faced the same friction.^[16] Between 2017 and 2019, Lambda pivoted away from computer vision software and into AI hardware and GPU infrastructure.^[1] Michael Balaban told The Next Platform that by late 2016 and early 2017 the founders realized many other AI teams faced the same AWS costs, and that selling their own servers online was "a better business than that AI application we had developed."^[16]

The company introduced its first dedicated deep learning workstations and GPU servers during this period, including the Lambda Quad workstation and Lambda Blade server.^[1] It later launched the Lambda Echelon (rolled out in late 2020), a turn-key on-premises GPU cluster that shipped pre-configured with compute, storage, and networking in rack form.^[16] Lambda opened its GPU cloud service publicly in 2018, making it one of the first cloud platforms dedicated specifically to deep learning workloads.^[1] The Lambda Stack, a curated suite of pre-installed deep learning software, launched alongside the cloud service to reduce setup time for practitioners.^[9]

Lambda's early customer base in this period included academic groups at universities such as Stanford, MIT, and Caltech, as well as a growing list of research labs at large technology companies.^[1] The company's positioning emphasized that researchers could rent the same kind of multi-GPU hardware they would otherwise have to purchase outright, and that the curated Lambda Stack removed many days of CUDA, driver, and framework debugging from a typical setup.^[9]

Growth and cloud focus (2020 to 2024)

Through 2020 to 2023, Lambda expanded its cloud GPU catalog to cover a broader range of NVIDIA architectures, including A100 and H100 GPUs, laying the groundwork for multi-node distributed training products such as 1-Click Clusters, which launched in 2024.^[1][11] The company reported growing adoption by universities, government research labs, and frontier AI companies during this period.^[1] Lambda's cloud revenue overtook its on-premises hardware business during this stretch, although hardware continued to be a meaningful contributor through 2024.^[14]

In June 2024, Lambda launched 1-Click Clusters, on-demand clusters of 16 to 512 NVIDIA H100 GPUs with NVIDIA Quantum-2 InfiniBand networking and a one-week reservation minimum, lowering the barrier for teams to run distributed training without negotiating long-term contracts or managing InfiniBand configuration themselves.^[11] Lambda argued that until then, short-term access to GPU clusters in the cloud had been available only to the largest enterprises.^[11]

Despite strong growth, Lambda faced repeated criticism in 2024 for GPU availability shortfalls. One developer reported that instances were often unavailable on demand during 2024, and documented a same-day A100 provisioning success rate of about 64 percent over a six-month window.^[19] The shortages reflected industry-wide constraints on advanced NVIDIA supply during the 2023 to 2024 generative AI boom, but they also strained Lambda's reputation among small developers who could not get reserved capacity.^[19]

Rebranding and expansion (2025 to present)

On March 25, 2025, Lambda announced that its primary web domain was migrating from lambdalabs.com to lambda.ai, reflecting the company's legal name, Lambda, Inc., and its positioning as a pure AI infrastructure provider.^[8] The former domain continued to redirect traffic during the transition period.^[8]

In August 2025, Lambda formally ended sales and support for its on-premises hardware product lines, including the Vector, Vector One, and Vector Pro workstations as well as the Scalar and Hyperplane servers. Customers with existing hardware remained covered by their original warranty terms. The decision concentrated the company's resources on cloud services and reflected a strategic shift toward operating its own data center capacity rather than shipping boxes to customer premises.

In September 2025, NVIDIA signed a $1.5 billion agreement to lease back 18,000 GPUs from Lambda over four years, making NVIDIA, at the time, Lambda's largest single customer.^[4] The deal, first reported by The Information and relayed by Data Center Dynamics and other outlets, was structured as approximately $1.3 billion for 10,000 higher-end GPUs and $200 million for an additional 8,000 units.^[4] Tom's Hardware reported that the leased servers were ones Lambda had bought from NVIDIA and would be used in-house by NVIDIA's researchers, and called the arrangement "nothing unusual" given NVIDIA's support for companies that use its products, noting a similar deal with CoreWeave.^[26]

Also in September 2025, Lambda and ECL brought the first hydrogen-powered NVIDIA GB300 NVL72 systems online at ECL's Mountain View campus, a zero-water and zero-emissions modular facility powered entirely by hydrogen fuel cells.^[18] The deployment was notable both for the chip generation, Blackwell Ultra, and for the unusual energy source.^[18] The Supermicro-built GB300 NVL72 racks consumed roughly 142 kilowatts each and used direct-to-chip liquid cooling.^[18]

In October 2025, Lambda announced plans to establish a major AI factory in Kansas City, Missouri.^[22] At announcement, the facility was expected to launch in early 2026 with 24 megawatts of capacity and the potential to scale beyond 100 megawatts over time.^[22] The Kansas City site, an unoccupied 2009-built facility, was to house more than 10,000 Blackwell Ultra GPUs at launch and is operated as a single-tenant supercomputer for a multi-year Lambda customer.^[23] Local media reported the investment at about $500 million.^[49]

In November 2025, Lambda announced a multibillion-dollar, multi-year agreement with Microsoft to deploy AI infrastructure powered by tens of thousands of NVIDIA GPUs, including GB300 NVL72 systems.^[13] TechCrunch reported that the exact size of the deal was not disclosed.^[3] Lambda and Microsoft had been collaborating in some form for more than eight years before the November 2025 announcement, but the new contract represented a step change in scale.^[13] "It's great to watch the Microsoft and Lambda teams working together to deploy these massive AI supercomputers," Stephen Balaban said. "We've been working with Microsoft for more than eight years, and this is a phenomenal next step in our relationship."^[13]

Later in November 2025, Lambda announced a Series E funding round of more than $1.5 billion led by TWG Global, with participation from US Innovative Technology Fund and other investors.^[7] Sacra Research put the round's post-money valuation at approximately $5.9 billion.^[14] Announcing the raise, Stephen Balaban said the financing "helps enable Lambda to develop gigawatt-scale AI factories that power services used by hundreds of millions of people every day."^[31] In January 2026, Data Center Dynamics, citing The Information, reported that Lambda was in talks to raise about $350 million in pre-IPO funding, with Mubadala Capital in talks to lead.^[42] Later reports described much larger rounds: up to $3 billion at a valuation of $12 billion or more in August 2026, then up to $4 billion at a $14.5 billion pre-money valuation, led by Coatue Management and Blackstone, in October 2026.^[14][40]

By early 2026, Lambda had announced plans for gigawatt-scale AI factories and added support for NVIDIA GB300 NVL72 and Vera Rubin NVL72 Superclusters.^[12] At NVIDIA GTC 2026, Lambda announced bare-metal instances on NVIDIA Vera Rubin NVL72, a production-scale GB300 NVL72 Supercluster using NVIDIA Quantum-X Photonics co-packaged optics, and one of the largest planned deployments of Quantum-X InfiniBand switches to date, in an AI factory containing more than 10,000 GB300 GPUs.^[24]

In May 2026, the company announced a major leadership reorganization.^[17] Michel Combes, formerly president of SoftBank Group International and CEO of Sprint, was named Lambda's chief executive officer.^[21] Co-founder Stephen Balaban transitioned to chief technology officer, where he would lead technology strategy and product full time.^[21] Co-founder Michael Balaban became chief product officer.^[17] John Donovan, the former CEO of AT&T Communications, was named chairman of the board.^[17] Charles Fisher, previously CFO at Turo and a senior finance executive at Charter Communications, joined as CFO.^[17] Leonard Speiser, a co-founder and former CEO of Clover, had been appointed chief operating officer in January 2026.^[41] The May release also listed earlier additions: Jerry Hunter (a former COO of Snap and early AWS infrastructure leader) as vice chairman of compute delivery and David Connolly, formerly general counsel at Altice, as chief legal officer.^[17] Lambda said the new team was assembled to scale its AI infrastructure footprint to roughly 3 gigawatts of compute under management by 2030.^[17] "When I met Stephen and his team, and saw what they have built, I knew this was the opportunity I have been looking for," Combes said of joining the company.^[17]

Leadership

Following the May 2026 reorganization, Lambda's executive team and board leadership stood as follows.

The restructuring split co-founder roles between technology and product (Stephen and Michael Balaban) and brought in operating leaders from large telecommunications and consumer technology businesses to manage the scale-out toward a stated goal of 3 gigawatts of AI compute under management by 2030.^[17] Several commentators noted that Lambda's executive hiring profile resembled that of a near-IPO telecommunications or data center operator more than that of a traditional cloud startup.^[21]

How much funding has Lambda raised?

Lambda has raised approximately $2.36 billion in equity funding across five named rounds, plus additional debt facilities; a further pre-IPO round of up to $4 billion was reported to be in progress in October 2026.^[14][40]

In addition to equity rounds, Lambda secured a $500 million debt facility from Macquarie Group in April 2024 and a $275 million credit facility from JPMorgan in August 2025, providing capital to purchase and finance GPU inventory ahead of customer contracts.^[1] In 2026 the scale of its borrowing rose sharply: it upsized the JPMorgan-arranged facility to $1 billion in May,^[43] closed a broadly syndicated term loan on August 27, and on October 1 closed a $1.008 billion investment-grade delayed-draw term loan at a 6.78 percent fixed rate, backed by contracts with two investment-grade customers.^[44] Debt-financed GPU acquisition has been a common pattern across the neocloud category, with CoreWeave using similar structures to fund its inventory before customer cash flow materialized.

NVIDIA participated directly in the Series D, establishing a formal investor relationship that complemented the GPU supply agreements Lambda held.^[6] The Series E was led by TWG Global, an investment firm founded by Thomas Tull and Mark Walter (the Guggenheim Partners CEO and Los Angeles Lakers co-owner) and backed in part by Abu Dhabi's Mubadala Capital, which anchors TWG's roughly $15 billion AI-focused fund.^[7]

By late 2025, Lambda was in discussions with investment banks including Morgan Stanley, JPMorgan, and Citi about a potential initial public offering.^[28] Contrary Research put Lambda's valuation as of October 2025 at roughly 5.9 times its 2024 revenue, compared with CoreWeave's market capitalization of about 23.4 times its trailing revenue at the time.^[1] In January 2026 Lambda was reported to be targeting an IPO in the second half of 2026,^[42] but by October 2026 the listing had been pushed back to a planned 2027 IPO amid market uncertainty.^[40]

Products

Lambda Cloud

Lambda Cloud is the company's primary revenue-generating product, providing on-demand access to NVIDIA GPU instances with hourly (or per-minute) billing, no egress fees, and pre-installed Lambda Stack software.^[14] The platform supports SSH access, persistent filesystems, and integration with standard orchestration tools including Kubernetes, Slurm, and dstack.

Instance types range from older NVIDIA Quadro RTX 6000 and Tesla V100 cards through A6000, A10, A100, GH200, H100, and B200 configurations.^[20] All instances include persistent storage and come pre-configured with the Lambda Stack deep learning environment.^[9]

Lambda Cloud is built around an asset-heavy model: the company owns the GPUs it rents out rather than reselling hyperscaler capacity.^[14] The company says its workstations and servers have been deployed in 97 percent of US universities and that more than 50,000 machine learning teams globally use its software stack.^[1] Sacra Research describes cloud GPU rental as the bulk of Lambda's revenue, with legacy hardware sales a shrinking portion.^[14]

1-Click Clusters

1-Click Clusters is Lambda's managed multi-node cluster product, designed for distributed AI training and large-scale inference.^[10] The product was introduced in its current form in mid-2024, following earlier iterations that required more manual configuration.^[11]

Clusters are built on NVIDIA HGX B200 SXM6 or HGX H100 nodes interconnected with NVIDIA Quantum-2 InfiniBand networking and SHARP (Scalable Hierarchical Aggregation and Reduction Protocol) acceleration, providing 3,200 Gbps of aggregate bandwidth per eight-GPU node.^[1] Cluster sizes range from 16 GPUs up to 2,000 or more, with self-service provisioning available through the Lambda dashboard.^[10] Commitment terms run from two weeks to one year, with per-GPU hourly pricing declining at larger scales and longer terms.^[10]

Lambda markets the product with managed Kubernetes or Slurm orchestration, S3-compatible storage integration, and SOC 2 Type II security certification.^[10] The company positions 1-Click Clusters as a faster alternative to negotiated enterprise GPU contracts, emphasizing that clusters can be provisioned in minutes rather than weeks.^[11] Several frontier AI labs have used 1-Click Clusters for experimental training runs that did not justify the latency of negotiated reserved capacity.

Superclusters

Superclusters is Lambda's highest-tier product, introduced in 2025 for frontier AI training at scale.^[12] Superclusters use NVIDIA GB300 NVL72 and, announced for later in 2026, NVIDIA Vera Rubin NVL72 systems.^[12] Each GB300 NVL72 rack integrates 72 NVIDIA Blackwell Ultra GPUs and 36 NVIDIA Grace CPUs, with 37 TB of fast memory and 130 TB/s of NVLink Switch bandwidth per rack.^[12]

The product targets frontier model training workloads that require deterministic network behavior across tens of thousands of GPUs. Lambda and ECL brought the first hydrogen-powered GB300 NVL72 systems online in September 2025.^[18] The Microsoft infrastructure agreement announced in November 2025 included NVIDIA GB300 NVL72 systems.^[13]

At NVIDIA GTC 2026 in March, Lambda announced that it would begin deploying NVIDIA Vera Rubin NVL72 systems as bare-metal instances in the second half of 2026, following a January announcement that Vera Rubin would become the core building block of its Superclusters.^[24][25] The Vera Rubin platform features 72 Rubin GPUs and 36 Vera CPUs per rack, with Lambda citing up to 10 times higher token throughput per watt and one-tenth the cost per million tokens compared with NVIDIA Blackwell.^[25] Lambda also disclosed plans to incorporate NVIDIA Quantum-X Photonics co-packaged optics switches into a production-scale Supercluster, in what Lambda described as one of the largest deployments of the new optical interconnect to date.^[24]

Private cloud

For enterprises with compliance, data sovereignty, or air-gap requirements, Lambda offers dedicated single-tenant GPU clusters that are physically isolated from shared infrastructure. Private cloud deployments typically start at 1,000 or more GPUs and are priced through direct sales engagement. Lambda targets regulated industries including finance, healthcare, pharmaceutical research, aerospace, and defense, as well as U.S. government agencies. The Kansas City AI factory announced in late 2025 is the largest known single-tenant private cloud deployment in Lambda's history, with more than 10,000 Blackwell Ultra GPUs dedicated to a single customer.^[23]

Lambda Inference API

Lambda launched a serverless LLM inference API offering OpenAI-compatible endpoints for popular open-weight models. The service was positioned on low per-token pricing and claimed to be among the lowest-cost inference options in the market at launch. Supported models included the Meta Llama 3.3 and Llama 4 series, Alibaba Qwen3, and DeepSeek models, among others. Pricing was structured around token consumption, with costs starting as low as $0.02 per million tokens for smaller models and reaching up to roughly $0.90 per million tokens for larger ones.

By early 2026, Lambda began transitioning away from the standalone Inference API product, directing customers toward deploying models on Lambda GPU instances instead.^[45] The inference endpoint product served as an entry point for developers before they scaled onto dedicated or cluster compute.

Lambda Stack

Lambda Stack is a curated set of deep learning software packages that Lambda maintains and tests for compatibility across its hardware.^[9] The stack is pre-installed on all Lambda cloud instances and on-premises systems.^[9] It includes:

• PyTorch (v2.7.0 as of early 2026)
• TensorFlow (v2.19.0)
• Keras (v3.10.0)
• JAX (v0.6.0)
• Triton (v3.3.0)
• CUDA Toolkit (v12.8.93)
• NCCL library (v2.26.2)
• NVIDIA Container Toolkit (v1.18.1)

Lambda tests each component for interoperability across its GPU catalog, including the latest HGX B200 and H200 SXM configurations.^[9] Lambda Stack can also be installed on non-Lambda Ubuntu systems via a publicly available install script, making it usable by researchers who own their own GPU hardware.^[9] Updates are delivered through standard Ubuntu package management.^[9]

GPU inventory

The following table summarizes the on-demand GPU instance types Lambda listed on lambda.ai/pricing in September 2026.^[20][32] Prices are per GPU-hour, exclude sales tax, VAT, or GST, and rise as the instance gets smaller (for example, H100 SXM at $3.99 on an eight-GPU instance and $4.29 on a one-GPU instance).^[32]

1-Click Cluster pricing (per GPU per hour, for terms of two weeks to one year, as listed in September 2026) is as follows:^[32]

Reserved commitments of one year or more are available at custom pricing through Lambda's sales team. Supercluster bare-metal pricing for GB300 NVL72 and Vera Rubin NVL72 systems is custom and contracted directly with Lambda's enterprise team. Lambda has not published a public rate card for Superclusters.

Data centers and infrastructure

Lambda's data center footprint has grown rapidly since 2024. As of late 2026, Lambda operated, contracted, or had announced capacity in the following regions:

Lambda uses a mix of self-developed sites and partnerships with data center operators such as Aligned, Cologix, ECL, EdgeConneX, and Prime Data Centers, where it leases 21 MW at the LAX01 campus in Vernon, California.^[22][47] In September 2026 it announced plans for a new data center at MidAmerica Industrial Park in Mayes County, Oklahoma.^[48] The company has publicly committed to scaling toward roughly 3 gigawatts of contracted power capacity by 2030, with gigawatt-scale individual sites planned for the late 2020s.^[17] The Microsoft agreement and the NVIDIA leaseback together account for a substantial share of forward-contracted capacity, although Lambda has not disclosed exact allocations.

Lambda has emphasized direct-to-chip liquid cooling for its Blackwell generation deployments. The Supermicro-built GB300 NVL72 racks at Lambda sites typically pull about 142 kilowatts each, which is well above the densities supported by traditional air-cooled facilities.^[18] The company has also positioned itself as an early adopter of advanced networking, including NVIDIA Quantum-2 InfiniBand for the H100 and B200 generation and Quantum-X Photonics for upcoming Vera Rubin deployments.^[24]

Competitive landscape

Lambda competes in the GPU cloud and AI infrastructure market against hyperscalers (AWS, Google Cloud Platform, Microsoft Azure), other GPU-native cloud providers, and on-premises GPU server vendors. Independent rating systems such as SemiAnalysis's ClusterMAX placed Lambda in the upper tier of GPU-native providers but consistently behind CoreWeave, which has been rated the only "Platinum" tier provider in several 2025 and 2026 assessments.^[27]

How does Lambda compare with other GPU-native cloud providers?

The table below compares Lambda with three major GPU-native cloud competitors as of early 2026.

CoreWeave is Lambda's closest direct competitor at scale. CoreWeave went public in March 2025 at an initial market capitalization of about $23 billion (roughly $68 billion by October 2025)^[1] and has aggressively secured NVIDIA GPU supply through similar leaseback and partnership arrangements. CoreWeave targets enterprise customers and hyperscalers with strong SLA guarantees and Kubernetes-native orchestration but carries higher list prices than Lambda for comparable GPU configurations. CoreWeave is the largest single neocloud by revenue.

RunPod competes primarily on price, offering spot instances and a marketplace model that can produce significantly lower per-hour costs for tolerant workloads. RunPod's flexibility (spot pricing, consumer-grade GPUs, serverless functions) attracts cost-sensitive developers and smaller teams, but the platform offers fewer enterprise features, and its on-demand multi-node clusters scale only to 64 GPUs.^[34]

Crusoe positions itself around sustainable infrastructure, using stranded natural gas and other waste energy sources to power GPU clusters. Crusoe listed on-demand H100 access at $3.90 per GPU-hour in September 2026, slightly below Lambda's $3.99 to $4.29 range.^[32][35]

The hyperscaler comparison is more complex: AWS, Google, and Azure all offer GPU compute, but with higher list prices, more complex pricing structures, and egress fees that can significantly increase total cost for large data-transfer workloads. Lambda's flat pricing with no egress fees is a frequently cited reason developers choose the platform over cloud giants for training and fine-tuning workloads. The Microsoft agreement signed in November 2025 partially blurs this distinction: Microsoft is simultaneously a competitor (via Azure's own GPU instances) and a wholesale customer of Lambda's capacity, an arrangement that mirrors CoreWeave's relationship with the same hyperscaler.^[3]

How does Lambda compare with the hyperscalers?

The major cloud providers all offer GPU compute, but they package it differently from neoclouds like Lambda. Hyperscalers bundle compute with proprietary managed services (storage, networking, identity, machine learning platforms) and charge premium prices for GPU instances. They also typically include egress fees, which can add substantial cost to data-intensive AI training pipelines.

Hyperscalers retain advantages in geographic coverage, managed services integration (databases, identity, observability), and procurement processes that align with how large enterprises buy cloud. Lambda's pitch to those enterprises is that for the GPU compute layer specifically, a neocloud provides better unit economics, faster provisioning, and access to the latest accelerators sooner.

Who uses Lambda?

Lambda's disclosed customer base spans research universities, AI startups, and large enterprises. Publicly confirmed customers or users have included Apple, MIT, Stanford University, Harvard University, Caltech, Kaiser Permanente, Tencent, the U.S. Department of Defense, and Microsoft (under the 2025 infrastructure agreement).^[1] Tom's Hardware reported in September 2025 that Lambda rented capacity to Microsoft and Amazon and was "in line to serve" Google, OpenAI, xAI, and Anthropic.^[26] In late August 2026, Anthropic signed a cloud deal with Lambda reported to be worth $35 billion, covering about 350 megawatts at a Hut 8-developed data center in Nueces County, Texas, according to a Reuters source; the companies did not immediately comment.^[46]

At the time, the September 2025 NVIDIA leaseback deal made NVIDIA Lambda's largest single customer, with NVIDIA using Lambda's infrastructure to run its own GPU-intensive workloads.^[4] By late 2025, multiple frontier AI labs had used Lambda clusters for some portion of their training or inference.

Common use cases for Lambda's cloud include:

• Large language model pretraining and continued pretraining, particularly for teams running multi-node distributed training on H100 or B200 clusters
• Fine-tuning and instruction tuning of open-weight models such as Meta's Llama series
• AI inference workloads requiring dedicated GPU capacity rather than shared serverless infrastructure
• Computer vision and image generation model training
• Academic and research computing at universities that lack on-premises GPU infrastructure
• Government and defense AI programs requiring secure, single-tenant compute environments
• Reinforcement learning experiments using interactive multi-GPU clusters
• Robotics simulation and policy training at AI-first robotics startups

Lambda's Lambda Stack has been adopted by more than 50,000 machine learning teams globally, including research groups that install it on their own hardware rather than using Lambda Cloud.^[1]

How does Lambda make money?

Lambda runs an asset-heavy business model: it raises debt and equity, purchases NVIDIA accelerators in volume, signs long-term power and colocation contracts, and rents the resulting capacity to AI customers.^[14] The economics of the model resemble a data center operator more than a software company. Several characteristics define Lambda's economic profile.

• Capital intensity. Each Blackwell-class GPU costs roughly $30,000 to $40,000 at distribution prices, and a single GB300 NVL72 rack with 72 GPUs and supporting infrastructure can cost several million dollars before facility and power costs. Building a 100-megawatt AI factory at Blackwell densities can require well over $1 billion in equipment and infrastructure.
• Debt-financed growth. Lambda funds GPU acquisition partly through asset-backed debt, with the GPUs themselves serving as collateral. The Macquarie and JPMorgan facilities established this pattern through 2024 and 2025.^[1]
• Reserved-revenue model. A large share of Lambda's revenue is contracted on multi-year reservations, which provides revenue visibility but also locks the company into specific configurations and price points.^[14]
• Cost-of-revenue dominated by depreciation, power, and rent. GPU depreciation is the largest single cost component. Power and colocation rent are the other major costs, with high-density racks magnifying the per-square-foot cost.
• Gross margins. Sacra Research estimated Lambda's gross margin at roughly 50 percent across all products in 2025, with the cloud business alone running closer to 61 percent excluding non-cloud lines.^[14]

Sacra Research estimated $425 million in 2024 revenue, about $700 million in calendar 2025 revenue, and a $760 million annualized run rate by December 2025; it also noted an August 2026 media report that Lambda expected more than $1.5 billion in 2026 revenue, a management projection.^[14] Contrary Research valued Lambda at approximately 5.9 times its 2024 revenue in October 2025, compared to CoreWeave's 23.4 times trailing revenue, suggesting the private market assessed Lambda at a significant discount to its publicly traded competitor.^[1]

Reception

Lambda has received broadly positive reviews from the developer community for pricing transparency, the absence of egress fees, and the quality of the Lambda Stack pre-configuration. SemiAnalysis's ClusterMAX 2.0 rating placed Lambda in the Silver tier as of late 2025, behind CoreWeave (Platinum) and Crusoe and Fluidstack (Gold) but ahead of many smaller specialists.^[27]

The primary recurring criticism in 2024 and into 2025 was GPU availability. Users documented frequent "temporarily unavailable" status for popular instance types, with some multi-GPU H100 configurations impossible to provision on short notice during peak demand periods.^[19] One developer writing on Medium described a 26-hour wait in late 2024 when trying to scale an H100 reservation from two to four GPUs.^[19] Lambda addressed availability issues partly through the expansion of its cluster inventory and longer-commitment reservation products, which improved reported availability for dedicated cluster customers even as on-demand availability remained unpredictable.

SemiAnalysis's ClusterMAX 2.0 review also said Lambda came in first among rated clouds for customer complaints.^[27] Lambda's response has been to emphasize its 1-Click Clusters and dedicated reservation paths, where single-tenant hardware reduces the variability associated with shared on-demand pools.

On the business side, Lambda's revenue growth attracted analyst attention. The 79 percent year-over-year run-rate growth reported in late 2025 outpaced most large cloud providers, although Lambda's absolute revenue remained much smaller than CoreWeave's.^[14] Analysts at PM Insights and Forge Global noted that secondary share prices for Lambda climbed in the months leading up to and following the Series E, reflecting investor enthusiasm for the IPO narrative.^[29]

Limitations

Several structural limitations were noted by analysts and users as of 2025 and early 2026:

GPU availability on on-demand instances remains constrained during periods of peak demand.^[19] Unlike hyperscalers with vast reserved capacity across dozens of regions, Lambda's footprint is smaller, and popular configurations can sell out quickly. Teams with hard deadlines or time-sensitive training runs often prefer reserved cluster contracts over on-demand access.

Lambda does not offer spot instances (preemptible compute) as of early 2026, a feature that RunPod and AWS both provide and that can reduce costs significantly for fault-tolerant workloads.

Geographic diversity is limited compared to hyperscalers. Lambda operates data centers in the United States and, as of early 2026, had not published a multi-region deployment map comparable to AWS or Google Cloud. This limits latency optimization for users outside North America and creates data residency constraints for international enterprise customers.

Lambda's heavy reliance on NVIDIA creates supply-chain risk. If NVIDIA were to reduce allocations, raise prices, or change partnership terms, Lambda's cost structure and GPU availability would be directly affected. CoreWeave faces the same dependency, but with a larger and more diversified inventory. Lambda has not publicly announced support for accelerators from AMD, Intel, or other alternative chip vendors as of mid-2026, although the company has stated that it evaluates the broader silicon landscape.

Customer concentration is also a risk. The September 2025 NVIDIA leaseback agreement made NVIDIA Lambda's largest single customer, and the November 2025 Microsoft deal added another large concentrated relationship.^[4] While these contracts provide revenue visibility, they expose Lambda to risk if either counterparty alters scope or timing.

The wind-down of the standalone Inference API product in early 2026 removed a low-friction entry point for developers who wanted to pay only for tokens rather than manage full GPU instances.^[45] While Lambda directs these users to its instance marketplace, the shift increases the minimum cost threshold for small-scale inference workloads.

Finally, Lambda's economics depend on continued capital availability for asset-heavy data center buildouts. If the broader market reassesses neocloud unit economics, valuations and debt terms could tighten quickly, as happened to several smaller GPU clouds during 2024.

Outlook

Lambda's near-term outlook is shaped by three converging dynamics. First, the Series E financing, the Microsoft contract, the NVIDIA leaseback, and the planned Kansas City AI factory all suggest the company is moving from a developer-focused on-demand GPU cloud into a wholesale infrastructure operator that supplies a small number of very large customers alongside its long tail of researchers. Second, Lambda's stated goal of roughly 3 gigawatts of contracted power by 2030 implies a multi-year capital expenditure program comparable in scope to mid-sized data center REITs.^[17] Third, the leadership reorganization in May 2026, which brought in operating executives from telecommunications and large-scale technology, signals that Lambda's board is preparing the company for public-market scrutiny and operational scale.^[21]

The public market reception of CoreWeave's 2025 IPO, alongside subsequent volatility in neocloud valuations, will likely shape Lambda's IPO timing and pricing. By October 2026, Lambda was reported to be raising up to $4 billion at a $14.5 billion pre-money valuation ahead of an IPO planned for 2027.^[40] Analysts at Forge Global and PM Insights have argued that a successful Lambda IPO could push the company's market capitalization well above the Series E valuation of $5.9 billion, although the comparison is sensitive to whether public investors continue to value GPU clouds on revenue multiples close to CoreWeave's.^[28]

Longer term, Lambda's strategic position depends on whether AI compute demand continues to outpace supply, whether NVIDIA maintains its dominance over the accelerator market, and whether power availability remains the binding constraint on data center growth. If any of these dynamics shift, Lambda will have to adapt its asset-heavy model to changes in either the chip stack or the energy stack underneath it.

Continue reading

• NvidiaNvidia Corporation is an American technology company that designs graphics processing units (GPUs) and is the world's dominant supplier of hardware for artificial intelligenceMentioned 56 times above
• CoreWeaveCoreWeave, Inc. is an American artificial intelligence cloud computing company headquartered in Livingston, New Jersey, that rents large clusters of NVIDIA GPUs to AI developers and enterprises and builds the…Mentioned 17 times above
• Deep LearningDeep learning is a subset of machine learning that uses artificial neural networks with multiple layers to automatically learn representations of data at multiple levels of abstractionMentioned 8 times above
• NNeocloudA neocloud is a cloud computing company whose business is renting out GPU capacity for artificial intelligence workloads, rather than selling the broad catalogue of storage, database, networking and…Mentioned 7 times above
• CrusoeCrusoe is an American AI infrastructure company headquartered in Denver, Colorado, that develops and operates data centers for artificial intelligence workloads, rents GPU capacity through its Crusoe Cloud…Mentioned 6 times above
• RunPodRunPod is a globally distributed GPU cloud computing platform that provides on-demand and serverless GPU compute for artificial intelligence training, fine-tuning, and inference.Mentioned 5 times above

More from See also

• NVIDIA H100
• NVIDIA B200
• NVIDIA Blackwell
• LLaMA
• Together AI
• Fireworks AI

Topics:AI CompaniesAI ComputeAI Infrastructure

References

1. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14 ^15 ^16 ^17 ^18 ^19 ^20 ^21 ^22Contrary Research. "Lambda Business Breakdown & Founding Story." research.contrary.com/company/lambda
2. TechCrunch. "AI data center provider Lambda raises whopping $1.5B after multibillion-dollar Microsoft deal." November 18, 2025. techcrunch.com
3. ^1 ^2TechCrunch. "Lambda inks multibillion-dollar AI infrastructure deal with Microsoft." November 3, 2025. techcrunch.com
4. ^1 ^2 ^3 ^4Data Center Dynamics. "Nvidia signs $1.5bn deal to lease its GPUs back from Lambda." September 2025. datacenterdynamics.com
5. Data Center Dynamics. "Lambda Labs closes $320m Series C funding round." February 2024. datacenterdynamics.com
6. ^Tech Funding News. "NVIDIA-backed Lambda lands $480M at $4B valuation to scale its AI cloud." February 2025. techfundingnews.com
7. ^1 ^2SuperbCrew. "Lambda Raises Over $1.5 Billion In Series E Funding Led By TWG Global." November 2025. superbcrew.com
8. ^1 ^2 ^3 ^4Lambda. "lambdalabs.com is now lambda.ai." March 25, 2025. lambda.ai/blog
9. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8Lambda. "Lambda Stack AI Software for Deep Learning & Machine Learning." lambda.ai/lambda-stack-deep-learning-software
10. ^1 ^2 ^3 ^4Lambda. "1-Click Clusters." lambda.ai/1-click-clusters
11. ^1 ^2 ^3 ^4 ^5Lambda. "Introducing Lambda 1-Click Clusters, a new way to train large AI models." lambda.ai/blog
12. ^1 ^2 ^3 ^4Lambda. "Gigawatt-Scale AI Factories: NVIDIA GB300 NVL72 System on Lambda Cloud." lambda.ai/blog
13. ^1 ^2 ^3 ^4Lambda. "Lambda Announces Multibillion-Dollar Agreement With Microsoft." BusinessWire. November 3, 2025.
14. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14Sacra Research. "Lambda Labs revenue, valuation & funding." sacra.com/c/lambda-labs
15. GPUCloudList. "Lambda Labs Review 2026: GPU Cloud Pricing, Performance & Verdict." gpucloudlist.com
16. ^1 ^2 ^3 ^4 ^5 ^6 ^7Next Platform. "The Serendipitous AI System And Cloud Builder." December 21, 2020. nextplatform.com
17. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11BusinessWire. "Lambda Assembles Leadership Team to Power Gigawatt-Scale AI Infrastructure for the Superintelligence Era." May 5, 2026.
18. ^1 ^2 ^3 ^4 ^5BusinessWire. "Lambda and ECL Bring the First Hydrogen-Powered NVIDIA GB300 NVL72 Systems Online." September 23, 2025.
19. ^1 ^2 ^3 ^4 ^5Medium. "Why I Stopped Using Lambda Labs for GPU Cloud." velinxs. medium.com
20. ^1 ^2ComputePrices. "Lambda Labs GPU Pricing: Compare 10+ GPUs." computeprices.com/providers/lambda
21. ^1 ^2 ^3 ^4Data Center Dynamics. "Lambda hires Michel Combes as CEO, co-founder Stephen Balaban shifts to CTO role." 2026. datacenterdynamics.com
22. ^1 ^2 ^3Data Center Dynamics. "AI cloud firm Lambda targets data center deployment in Kansas City, Missouri." October 2025. datacenterdynamics.com
23. ^1 ^2Lambda. "Lambda Doubles Down on Midwest Expansion, To Build AI Factory in Kansas City, MO." October 28, 2025. lambda.ai/blog
24. ^1 ^2 ^3 ^4Lambda. "Lambda at NVIDIA GTC 2026: building the Superintelligence Cloud." lambda.ai/blog
25. ^1 ^2Lambda. "NVIDIA's Vera Rubin NVL72 coming to Lambda's Superintelligence Cloud." lambda.ai/blog
26. ^1 ^2Tom's Hardware. "Nvidia signs $1.5 billion deal with cloud startup Lambda to rent back its own AI chips." 2025. tomshardware.com
27. ^1 ^2 ^3SemiAnalysis. "ClusterMAX 2.0: The Industry Standard GPU Cloud Rating System." newsletter.semianalysis.com
28. ^1 ^2Forge Global. "Insights: Lambda Upcoming IPO & Private Stock Price." forgeglobal.com
29. ^PM Insights. "Lambda Labs Secondary Shares Climb 12% in the Last 90 Days With IPO Speculation Building." pminsights.com
30. Crescent Cove. "Crescent Cove Conversations featuring Stephen Balaban, Co-Founder & CEO, Lambda." 2024. crescentcove.com
31. ^Lambda. "Lambda Raises Over $1.5B From TWG Global, USIT to Build Superintelligence Cloud Infrastructure." November 18, 2025. lambda.ai/blog
32. ^1 ^2 ^3 ^4 ^5 ^6Lambda. "AI cloud pricing." Official pricing page, accessed September 28, 2026. lambda.ai/pricing
33. ^CoreWeave. "CoreWeave Cloud Pricing." Official pricing page, accessed September 28, 2026. coreweave.com/pricing
34. ^1 ^2Runpod. "GPU Cloud Pricing." Official pricing page (labeled "Updated September 27, 2026"), accessed September 28, 2026. runpod.io/pricing
35. ^1 ^2Crusoe. "GPU pricing for AI compute & inference." Official pricing page, accessed September 28, 2026. crusoe.ai/...pricing
36. ^Amazon Web Services. EC2 On-Demand price data, US East (N. Virginia), Linux (file publication date September 25, 2026), accessed September 28, 2026. b0.p.awsstatic.com/...Linux
37. ^Google Cloud. "Accelerator-optimized VM pricing." Official pricing page, accessed September 28, 2026. cloud.google.com/...accelerator-optimized
38. ^Microsoft. Azure Retail Prices API, Virtual Machines H100 SKUs in East US, accessed September 28, 2026. prices.azure.com/...prices
39. ^Runpod Documentation. "Pricing" (Pods). Accessed September 28, 2026. docs.runpod.io/...pricing
40. ^1 ^2 ^3 ^4 ^5TechCrunch. "AI computing startup Lambda to raise $4B ahead of planned IPO." October 6, 2026. techcrunch.com/...to-raise-4b-ahead-of-planned-ipo
41. ^Lambda. "Lambda Appoints Leonard Speiser as Chief Operating Officer." BusinessWire, January 8, 2026. businesswire.com/...ser-as-Chief-Operating-Officer
42. ^1 ^2Data Center Dynamics. "Lambda in talks to raise $350m in pre-IPO funding - report." January 9, 2026. datacenterdynamics.com/...n-pre-ipo-funding-report
43. ^Lambda. "Lambda closes $1 billion senior secured credit facility to meet gigawatt-scale AI infrastructure demand." May 7, 2026. lambda.ai/...illion-senior-secured-credit-facility
44. ^Lambda. "Lambda closes $1 billion senior secured fixed rate financing." October 1, 2026. lambda.ai/...n-senior-secured-fixed-rate-financing
45. ^1 ^2Lambda. "Run inference your way" (Inference API wind-down notice). Accessed November 2026. lambda.ai/inference
46. ^Reuters (via The Star). "Anthropic signs $35 billion cloud deal with Nvidia-backed Lambda, source says." September 1, 2026. thestar.com.my/...nvidia-backed-lambda-source-says
47. ^Prime Data Centers and Lambda. "Prime Data Centers and Lambda Partner to Power the Next Era of Superintelligence with AI-Optimized Infrastructure in Southern California." November 13, 2025. lambda.ai/...infrastructure-in-southern-california
48. ^Lambda. "Lambda to build new data center in Mayes County, Oklahoma." September 27, 2026. lambda.ai/...new-lambda-data-center-in-oklahoma
49. ^KCTV5. "Tech company to bring $500 million supercomputer facility to Kansas City." October 28, 2025. kctv5.com/...on-supercomputer-facility-kansas-city

Show all 49 references