NVIDIA Developer Webinar
Getting Started with NVIDIA DGX Spark: The Personal AI Supercomputer
Generative AI models are growing in complexity, pushing the limits of traditional local development environments. Developers often face "Out of Memory" errors or struggle with fragmented software stacks, forcing premature moves to costly cloud resources for experimentation.
Join us for an exclusive look at NVIDIA DGX Spark, a new class of computer designed from the ground up to build and run AI. Powered by the NVIDIA GB10 Grace Blackwell Superchip, this compact system brings 1 petaFLOP of AI performance and 128GB of unified memory directly to your desktop.
In this session, we will explore how DGX Spark bridges the gap between local prototyping and data center deployment, enabling you to fine-tune massive models (like Llama 3.1 70B) and run complex agentic workflows locally before scaling out.
Register for this webinar and explore the benefits of NVIDIA DGX Spark to accelerate your AI workload. Time will be available at the end of the webinar for Q&A.
Who Should Attend
Developers with Python proficiency and a basic understanding of LLM theory who are working in generative AI and seeking to enhance their knowledge of LLM and personal accelerated compute.
Prerequisites
- Python programming proficiency
- Understanding of AI agents, LLMs
- Familiarity with GPU computing
Learnings
In this webinar, you'll gain a deep understanding of NVIDIA’s AI Enterprise stack and learn :
- The Hardware Advantage: Understand how the NVIDIA GB10 Grace Blackwell Superchip and 128GB of unified memory allow for training and inference of models up to 200B parameters locally.
- Unified Software Stack: Learn how DGX Spark uses the same NVIDIA AI software stack found in data centers, ensuring your code is portable and scalable from day one.
- Advanced Use Cases: Discover how to perform local fine-tuning of Large Language Models (LLMs) and Vision Language Models (VLMs) without hitting memory walls.
- Scalability: See how to connect multiple DGX Spark units via NVIDIA ConnectX-7 networking to tackle even larger workloads (up to 405B parameters).
Don't miss this opportunity to enhance your expertise or master a new technology.
