Deploying Open-Weight Models on AWS: A Builder Webinar
Open-weight models are one of the most exciting shifts in AI right now, and they're fast becoming a core part of how builders ship coding and agentic workloads.
In this Session, we will go through:
01/ Landscape of the open-weight model ecosystem
02/ Open-weight models on AWS
03/ Deploy & scale open-weight intelligence on AWS
04/ Live Q&A with AWS experts
Our Speakers are:
Ryan Tan: Solution Architect @AWS Startup Team
Xiaowei Zhu: International Business Development Lead @AWS (Founder of ShallWe.tech)
Dr Kaige Yang: Senior Applied Scientist @AWS
AWS is expanding its catalogue of open-weight models to bring more of these options to you, so you can put them to work alongside the leading frontier models you already run. Because the best AI stacks aren't built on one model, they're built on choice: matching the right model to each job.
Open-weight models bring real strengths to that mix: more flexibility to customise, more control over where and how a model runs, and a strong fit for many of the coding and agentic use cases teams are building today. This session is about what's available on AWS and how to deploy it.
It's a live, hands-on webinar for builders and developers who want a clear picture of the open-weight landscape on AWS and the practical options for getting a model into production. Held at L200 (concept plus how-to), you'll leave knowing what's available, how the deployment paths compare, and enough detail to stand up your first deployment. As we expand open-weight availability on AWS, we want to make sure builders know what's on offer and how to put it to work.
What we'll cover
The open-weight landscape on AWS: what open-weight models are, which ones are available, where they run, and how they fit alongside frontier models in your stack.
The main deployment options and how to choose between them: Amazon Bedrock (managed and Custom Model Import), Amazon SageMaker JumpStart and real-time endpoints, and self-managed on EC2/EKS with GPUs, with the trade-offs across cost, latency, operational burden, and control.
A live end-to-end demo: standing up an open-weight model (e.g. via SageMaker JumpStart) and calling it, so you can see the path from zero to a working endpoint.
Scaling and cost tips, resources, and how to go deeper, in an open Q&A.
Who should attend
Builders and developers who want to deploy open-weight models on AWS and are looking for a clear, practical starting point. No prior deployment experience with these models required. This is an accessible L200 session.
This is a free event. Spaces are limited.