In twelve months Lambda grew from one compute model to five. Event Functions (the classic model, renamed), Web Functions for HTTP servers billed on active CPU, Durable Functions for long workflows, MicroVMs for isolated sandboxes, and Managed Instances on EC2. What each is for and how each bills.
For ten years "a Lambda function" meant one thing: an event handler, one request per execution environment, billed on memory multiplied by duration. That model now has a name, Event Functions, and four siblings. Each one removes a reason teams used to leave Lambda.
Book a free 30-minute call. You'll leave knowing what we'd do first and roughly what it costs, whether or not you hire us.
Event Functions charge for wall-clock time, so a request that waits 150 ms on a database pays for the wait. Web Functions charge for active CPU and for memory the environment holds, and one environment serves many requests at once. AWS's own example: an I/O-heavy API with 10 million requests a month, 200 ms each of which 50 ms is CPU, costs about $19.61 a month on Arm. That is the billing model that made Cloudflare Workers and Vercel's Fluid Compute cheap for this shape of traffic, now inside your AWS account and VPC.
Coding agents, notebooks and "run this script" features need somewhere isolated to execute code a user or a model wrote. MicroVMs give each session its own Firecracker VM, launched from a pre-initialized snapshot rather than a cold boot, suspended with memory and disk intact when idle, and resumed when needed. AWS documents integrations as a sandbox for Claude Managed Agents, the OpenAI Agents API and Cursor's self-hosted cloud agents. At launch they run on Arm in five Regions, with up to 16 vCPUs and 32 GB of memory.
Reacting to an event (S3, SQS, EventBridge, a schedule): Event Functions.
Serving HTTP from a framework you already use: Web Functions, once your runtime is supported (Node.js at launch).
A process with waits, approvals or many steps: Durable Functions.
Running untrusted code per user or per agent session: MicroVMs.
A function that is busy all day and you hold EC2 Savings Plans: Managed Instances.
Once a month: what changed in tech that month and what it means for you, from cloud and AI to security and the tools you build with. Every item links to its source. No sales pitch, unsubscribe in one click.
Book a free 30-minute call about it. Bring your setup and your bill; you'll leave knowing what we'd do first and roughly what it costs, whether or not you hire us.
Or keep reading
Lambda Web Functions vs Event Functions: Where the New Model Saves Money, and Where It Doesn't
Web Functions bill active CPU and held memory; Event Functions bill wall-clock time multiplied by memory. For AWS's own three example workloads, classic Lambda at 512 MB is cheaper. Once requests spend seconds waiting (LLM streaming, slow APIs, long polls) Web Functions win by 40 times or more: a 10-second stream costs about $3 per million requests instead of $134. The arithmetic, the crossover and a checklist.
Read
Lambda Web Functions: Run Express or Hono on Lambda and Pay for Active CPU
Launched in October 2026, Web Functions run a normal HTTP server on Lambda: no adapter, many concurrent requests per environment, native streaming, and billing on requests, active vCPU and held memory. AWS's worked examples, how it compares with Event Functions and function URLs, and what to check before moving.
Read
As of October 25, 2026. Provided as is, for general information, and accurate to the best of our knowledge on the date shown. Not legal, compliance, tax, financial or other professional advice. Prices, limits and vendor services change; check current vendor documentation before relying on anything here. Terms of use.