# AWS Lambda Now Runs GPU Inference with 48 Concurrent Invocations Sharing One Model per GPU

Source: [InfoQ](https://www.infoq.com/news/2026/11/aws-lambda-gpu-managed-instances/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=global) · 2026-11-22 · Cloud & DevOps
Page: https://themorning.dev/story/aws-lambda-now-runs-gpu-inference-with-48-concurrent-invocations-sharing-one-mod-adf1e140fe9b58fd
Kind: launch · Impact: medium

## In brief

- AWS Lambda now runs Event Functions on EC2 G-series GPU instances through Lambda Managed Instances.
- Up to 48 concurrent invocations can share one model per GPU.
- vLLM and SGLang base images need no handler code, and asynchronous invocations get 90-minute timeouts.

## Why it matters

Developers can run GPU-backed model inference on Lambda with shared models and concurrency, without writing a custom handler when using the vLLM or SGLang base images.

## Coverage

- [AWS Lambda Now Runs GPU Inference with 48 Concurrent Invocations Sharing One Model per GPU](https://www.infoq.com/news/2026/11/aws-lambda-gpu-managed-instances/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=global) — InfoQ, 2026-11-22

Tags: aws, lambda, gpu, inference, serverless

---

Brief generated by AI (claude-haiku-5-5) from the reporting listed under Coverage. Headlines belong to their publishers. The Morning Dev – https://themorning.dev
