AWS Lambda adds support for GPU instances on Lambda Managed Instances

AWS Lambda GPU-Accelerated Event Functions
AWS Lambda now supports GPU-accelerated Event Functions on Lambda Managed Instances using EC2 G-series instances. You can run GPU-accelerated workloads such as AI inference, media processing, 3D rendering, scientific computing, and financial simulations on Lambda without managing GPU drivers, frameworks, or scaling.
Lambda scales responsively based on GPU utilization, CPU utilization, and concurrency to keep your GPUs busy and reduce costs. GPU-accelerated functions integrate with over 200 AWS event sources.
What to do
- Use preconfigured GPU base images with frameworks like vLLM, SGLang, and PyTorch.
- Utilize container images and NVMe ephemeral storage of up to 32 GB.
- Leverage Seekable OCI (SOCI) lazy loading and stream artifacts from Amazon S3.
- Select G-series instance types from multiple generations for resilience.
Available in all AWS Commercial Regions where Lambda Managed Instances and EC2 G-series instances are available. Configure GPU-accelerated functions using the AWS Console, AWS CLI, AWS SDKs, AWS CloudFormation, AWS CDK, or AWS SAM. You pay the EC2 On-Demand price plus a 9.75% Lambda management fee and $0.20 per million requests.
Source: AWS release notes
If you need further guidance on AWS, our experts are available at AWS@westloop.io. You may also reach us by submitting the Contact Us form.


