Events & Updates

Recent updates

Past events, research milestones, press, and system notices.

• Jul 13, 2026
  research milestone

  Scaling Run Completed: Hyperparameter Optimization on Up to 192 GPUs

  Professor Kay Giesecke’s team, in Management Science and Engineering, completed a scaling run on Marlowe across up to 24 nodes (192 H100 GPUs) for hyperparameter optimization ahead of a larger campaign. The run began Friday, July 10 at 3:00 PM...

• Jul 1, 2026
  research milestone

  First Hero Run Concludes; Next Cycle Opens July 2

  Next cycle begins Thursday, July 2, 2026, 10:00 AM PDT Marlowe’s first hero run has completed — the PSI-32B training campaign across 24 nodes (192 H100 GPUs). The model is now in benchmark evaluation. The next dedicated hero-run cycle is...

• Jun 3, 2026
  talks seminars

  NVIDIA & Marlowe: Scaling Code Agents for 100M+ LOC

  Wednesday, June 3, 2026 | 2:00 – 3:30 PM PDT CoDa Building, Room W101, 1st Floor Abstract This talk will cover how to use NVIDIA GPUs and GPU-accelerated techniques to efficiently index, update and search massive code-bases with over 100M...

• May 20, 2026
  talks seminars

  NVIDIA & Marlowe: Workload Performance and Health Monitoring in Data Centers

  Wednesday, May 20, 2026 | 2:00 – 3:30 PM PDT CoDa Building, Room W401, 4th Floor Abstract This session explores how to efficiently run and monitor large workloads in hyper-scalars and the kind of telemetry data we can monitor from...

• May 13, 2026
  research milestone

  First Hero Run Scheduled: PSI-32B Training Campaign on 192 GPUs

  Begins Wednesday, May 27, 2026, 10:00 AM PDT Marlowe’s first dedicated hero run is scheduled to begin Wednesday, May 27 at 10:00 AM PDT. Professor Dan Yamins’ team will hold a reservation across 24 nodes (192 H100 GPUs) for a...

• Apr 22, 2026
  talks seminars

  NVIDIA & Marlowe: Training Robots with World Foundation Models

  Wednesday, April 22, 2026 | 2:00 - 3:30 PM PDT CoDa Building, Room W101, 1st Floor Abstract This talk explores the new wave of robotics driven by world foundation models. We present our family of world models and describe how...

• Mar 11, 2026
  talks seminars

  NVIDIA New AI Models: Open Models — Open to Build

  Wednesday, March 11, 2026 | 2:00 - 3:30 PM PDT CoDa Building, Room W401, 4th Floor Abstract This talk will cover NVIDIA’s new wave of open AI models, data and tooling that enable researchers to build high-performance, domain-specialized systems. It...

• Feb 20, 2026
  maintenance

  Planned Maintenance Window: February 24-25

  Marlowe will undergo scheduled maintenance on February 24-25 for firmware updates and network fabric optimization. The maintenance window is expected to last approximately 8 hours. Running jobs will be checkpointed where possible, and the scheduling queue will drain starting February...

• Feb 18, 2026
  talks seminars

  NVIDIA & Marlowe: Post-training Language Agents with NeMo RL and NeMo Gym

  Wednesday, February 18, 2026 | 2:00 - 3:30 PM PST CoDa Building, Room W401, 4th Floor Abstract This talk covers modern post-training and reinforcement learning techniques used to train effective language models and agents. An overview of NVIDIA NeMo and...

• Feb 7, 2026
  research milestone

  Milestone: Training POC for 80B-Parameter Brain Model on Marlowe Completed

  Professor Dan Yamins’ team completed a proof-of-concept hero run on Marlowe, validating that an 80-billion-parameter brain-inspired neural network can train at scale across 24 nodes (192 H100 GPUs). The 24-hour run achieved 45% model FLOPS utilization (MFU), confirming that capability-scale...

• Jan 28, 2026
  talks seminars

  Compute Resources @ Stanford and Beyond

  Wednesday, January 28, 2026 | 2:00 - 3:30 PM PST CoDa Building, Room W401, 4th Floor, 389 Jane Stanford Way, Stanford, CA 94305 Abstract This session covers the foundational knowledge of GPUs, including their architecture, functionality, and applications in computing....

• Nov 12, 2025
  talks seminars

  NVIDIA PhysicsNeMo: Community Models and Dataset Integrations

  Wednesday, November 12, 2025 | 2:00 - 3:30 PM PST CoDa Building, E365 Conference Room, 3rd Floor Abstract Scientific machine learning (SciML) is becoming a fundamental part of research, development, and discovery workflows for scientists across many domains such as...

• Oct 29, 2025
  talks seminars

  Building Scalable, End-to-End Generative AI with NVIDIA NeMo Framework on Marlowe

  Wednesday, October 29, 2025 | 2:00 - 3:30 PM PDT CoDa Building, Rm E401, 4th Floor Abstract As foundation models grow in scale and complexity, the need for reproducible, modular, and high-performance research frameworks has become critical. NVIDIA NeMo is...

• Sep 24, 2025
  talks seminars

  Graph Neural Networks & LLMs in PyG on Marlowe

  Wednesday, September 24, 2025 | 2:00 - 3:30 PM PDT CoDa Building, Rm E401, 4th Floor Abstract This talk will cover how Graph Neural Networks can be used to enhance LLMs using PyG to improve accuracy for RAG-like tasks across...

• Aug 13, 2025
  talks seminars

  Using CuPyNumeric and the Legate Ecosystem for Multi-GPU Scaling on Marlowe

  Wednesday, August 13, 2025 | 2:00 - 3:30 PM PDT CoDa Building, Rm E401, 4th Floor Abstract Discover how to use CuPyNumeric for seamlessly scaling NumPy code to multi-GPU and multi-node setups. This workshop will discuss the best practices and...

• Aug 7, 2025
  education

  DataSci 211: Accelerating Research with Marlowe

  Stanford Data Science launched DATASCI 211: Accelerating Research with Marlowe — Practical GPU Computing for Scientists, a 1-unit course offered in Autumn quarter 2025. The course provides hands-on and conceptual grounding in high-performance GPU computing on Marlowe. Aimed at researchers...

• Jul 16, 2025
  talks seminars

  NVIDIA Clara for AI-Enabled Healthcare and Life Sciences on Marlowe

  Wednesday, July 16, 2025 | 2:00 - 3:30 PM PDT CoDa Building, Rm E401, 4th Floor Abstract This session provides an overview of NVIDIA Clara, which is an accelerated, AI-enabled platform for healthcare and life sciences. It will specifically highlight...

• Jun 25, 2025
  talks seminars

  NVIDIA Nsight Systems for Profiling Code on Marlowe

  Wednesday, June 25, 2025 | 2:00 - 3:30 PM PDT CoDa Building, Rm E401, 4th Floor Abstract Discover how to use Nsight Systems for system-wide profiling of GPU-accelerated applications. This workshop will help you identify performance bottlenecks and optimize code...

• May 21, 2025
  talks seminars

  Distributed Training & Marlowe Multi-GPU Best Practices

  Wednesday, May 21, 2025 | 2:00 - 3:30 PM PDT CoDa Building, Rm E401, 4th Floor Abstract This training focuses on efficient strategies for using multiple GPUs and nodes. We will overview how to deploy strategies of data parallelism and...

• Apr 29, 2025
  talks seminars

  Marlowe Featured at 2025 Stanford Data Science Conference

  Marlowe researchers presented a dedicated session at the 2025 Stanford Data Science Conference, held at the Simonyi Conference Center in CoDa on April 29, 2025. The session, Designing for Impact with Marlowe GPU-Based Computational Instrument, featured three talks showcasing research...

• Apr 23, 2025
  talks seminars

  Marlowe GPU Computing Foundations: Architecture, Applications, and Acceleration

  Wednesday, April 23, 2025 | 2:00 - 3:30 PM PDT CoDa Building, Rm E401, 4th Floor Abstract This session covers the foundational knowledge of GPUs, including their architecture, functionality, and applications in computing. It provides an introduction to GPU computing...

• Dec 15, 2024
  press

  Stanford Welcomes First GPU-Based Supercomputer

  Stanford Report published a feature story on Marlowe, Stanford’s first GPU-based supercomputer. The article covers Marlowe’s role in democratizing frontier AI research at the university, making capabilities previously available only to industry accessible to faculty across seven schools. Read the...

No updates in this category.