Lambda

Lambda CTO and co-founder Stephen Balaban joins Fortune Brainstorm Tech in Aspen this week. His session, "The AI value gap," asks the question enterprises are sitting with right now: AI budgets are up across the board, but the financial impact isn't matching the investment. Where are the structural barriers between ambition and operational results? Stephen joins Manoj Bohra (State Street), Bill Briggs (Deloitte), and Faraz Shafiq (Wells Fargo) for the conversation on June 10. Read more: https://lnkd.in/e8EczmFe #BrainstormTech

Lambda is collaborating with NVIDIA at the frontier of AI networking. As clusters scale to tens of thousands of GPUs, the network sits in the critical path of every token. Agentic workloads span model calls, tool calls, retrieval, and reasoning. More east-west traffic. More pressure on the fabric. Co-packaged optics moves optical conversion next to the switch ASIC. The electrical path drops from centimeters to micrometers. The DSP disappears. Power drops. Failure points follow. Lambda bet early on NVIDIA Quantum-X InfiniBand Photonics because the network is no longer background infrastructure. It's part of the compute envelope. Read more: https://lnkd.in/eq4TD2Qg

Lambda is proud to sponsor the 10th edition of RAAIS. Sam Khosroshahi and David Hartmann will be there. See you in London.

28 trucks. No humans in the cab. Kodiak trains GigaFusionNet, the model behind the Kodiak Driver, on Lambda’s NVIDIA HGX H100 clusters. The model fuses camera, LiDAR, and radar data into a single backbone that can handle everything from 3D object detection to end-to-end driving. To build the foundation model, Kodiak needed to scale fast without moving petabytes of sensor data. They were pre-training on Lambda within a week and running twice as many experiments at twice the speed. The foundation model can be further distilled on NVIDIA DRIVE Hyperion for smooth in-vehicle operations. Read the full story: https://lnkd.in/ezSSdiMp

"Trust everyone" isn't an access policy. Workspaces for Lambda Cloud give you resource-level access control and a clean way to group instances, storage, and users by project. Scope what your team can see and launch. Available on every Lambda Cloud account via UI. Read more: https://lnkd.in/ed8GNW47

NVIDIA announced DSX, Vera Rubin ramping into full production, Vera CPUs, and DSX OS at NVIDIA GTC Taipei. Lambda is investing across all of it. Agentic workloads are getting harder to run. These technologies close that gap. We're building the Superintelligence Cloud around them so your team can spend less time wrestling with infrastructure and more time on the work that matters. #GTCTaipei #agenticAI https://lnkd.in/gNtrhG8d

Lambda is first to adopt #NVIDIA co-packaged optics to make large-scale AI clusters more efficient, resilient, and ready for agentic AI. Large GPU clusters are becoming power-bound, reliability-bound, and network-bound. Lambda is taking an early look at co-packaged optics to reduce network overhead and improve tokens per watt. More compute in the same footprint. Fewer interruptions. Better infrastructure for agentic AI. Read more: https://lnkd.in/eSyAVN5y

Heading to #CVPR2026 in Denver? Save Saturday night. Lambda, NVIDIA, and Kodiak are hosting AI After Dark. The official program ends. The real conversations start. June 6. Researchers, engineers, and builders. Spots are limited; approval required. Register: https://luma.com/in9oubm7 #CVPR2026 #AIAfterDark

📣 NVIDIA Blackwell sets a new STAC-AI LANG6 record for LLM inference in quantitative research and algorithmic trading, delivering the highest compute-per-watt and lowest token cost. We tested Llama 3.1 8B and 70B with NVIDIA TensorRT-LLM across multiple NVIDIA platforms. Systems tested: ✅ NVIDIA HGX B200 on Lambda ✅ NVIDIA RTX PRO 6000 Blackwell Server Edition system from Supermicro ✅ NVIDIA Grace Hopper-based server from Hewlett Packard Enterprise See the results 👉 https://nvda.ws/4fFM5ww