Skip to main content
This guide walks you through deploying the AWS Lambda infrastructure needed to render videos in the cloud using Remotion Lambda.

​
Prerequisites

Before deploying to AWS Lambda, you need:
  • An AWS account with appropriate permissions
  • AWS access credentials (Access Key ID and Secret Access Key)
  • Node.js installed locally

​
Environment Variables

First, configure your AWS credentials in your .env file:
​
REMOTION_AWS_ACCESS_KEY_ID
string
required
Your AWS Access Key ID. This is used to authenticate with AWS services.
​
REMOTION_AWS_SECRET_ACCESS_KEY
string
required
Your AWS Secret Access Key. Keep this secure and never commit it to version control.
​
OPENAI_API_KEY
string
required
Your OpenAI API key for generating video content from prompts.

​
Deployment Process

The deployment script (deploy.mjs) handles the complete setup automatically:
1

Verify AWS Credentials

The script checks that your AWS credentials are properly configured:
if (!process.env.AWS_ACCESS_KEY_ID && !process.env.REMOTION_AWS_ACCESS_KEY_ID) {
  console.log('The environment variable "REMOTION_AWS_ACCESS_KEY_ID" is not set.');
  process.exit(0);
}
If credentials are missing, the deployment will exit with instructions to complete the Lambda setup.
2

Deploy Lambda Function

The script deploys a Lambda function with the configured resources:
const { functionName, alreadyExisted } = await deployFunction({
  createCloudWatchLogGroup: true,
  memorySizeInMb: RAM,
  region: REGION,
  timeoutInSeconds: TIMEOUT,
  diskSizeInMb: DISK,
});
This creates or updates the Lambda function with:
  • Memory: 3009 MB (configured in config.mjs)
  • Timeout: 240 seconds (4 minutes)
  • Disk: 10240 MB (10 GB)
  • Region: us-east-1 (configurable)
  • CloudWatch Logs: Automatically created for monitoring
3

Create or Get S3 Bucket

An S3 bucket is created to store rendered videos and deployment artifacts:
const { bucketName, alreadyExisted } = await getOrCreateBucket({
  region: REGION,
});
The bucket is automatically named and configured for the specified region.
4

Deploy Remotion Site

Your Remotion composition is bundled and deployed to the S3 bucket:
const { siteName } = await deploySite({
  bucketName,
  entryPoint: path.join(process.cwd(), "src", "remotion", "index.ts"),
  siteName: SITE_NAME,
  region: REGION,
  options: { webpackOverride },
});
This bundles your video templates and makes them available for Lambda rendering.

​
Running the Deployment

Execute the deployment script: 
node deploy.mjs
You’ll see output like: 
Selected region: us-east-1
Deploying Lambda function... remotionlambda-mem3009mb-disk10240mb-240sec (created)
Ensuring bucket... remotionlambda-useast1-xxxxxx (created)
Deploying site... my-next-app

You now have everything you need to render videos!

​
When to Re-deploy

Re-run the deployment command when you:
  1. Changed the video template or composition
  2. Modified config.mjs settings
  3. Upgraded Remotion to a newer version
The deployment script is idempotent - it will update existing resources rather than creating duplicates.

​
Next Steps

Once deployed, you can:

​
Troubleshooting

​
Missing Credentials

If you see an error about missing credentials, ensure your .env file contains: 
REMOTION_AWS_ACCESS_KEY_ID=your_access_key_here
REMOTION_AWS_SECRET_ACCESS_KEY=your_secret_key_here

​
Permission Errors

Your AWS IAM user needs permissions for:
  • Lambda (create/update functions)
  • S3 (create buckets, upload objects)
  • CloudWatch Logs (create log groups)
  • IAM (create roles for Lambda execution)
See the official Remotion Lambda setup guide for detailed IAM policy requirements.

Build docs developers (and LLMs) love

Get started for freeTalk to us