Run major GCP services locally
Start one Docker runtime, inspect its local services in the console, and work without a shared cloud development project for every iteration.
Browse the service catalogLocalCloud brings an all-in-one local-development pattern to Google Cloud: service guides, generated local SDK endpoints, and a built-in console. Its proprietary Public Preview License is free for individuals and organizations, including for-profit companies, for non-production development and testing workflows.
LocalStack is designed around AWS service emulation. LocalCloud serves a similar category for Google Cloud APIs under a proprietary Public Preview License. Run local GCP services for development and testing, then validate separately against real Google Cloud.
Swipe horizontally to compare all columns.
| Question | LocalStack | LocalCloud |
|---|---|---|
| Primary cloud | AWS workflows | Google Cloud workflows |
| Local development goal | Develop and test AWS-connected applications locally | Develop and test GCP-connected applications locally |
| LocalCloud fit | Not a substitute for AWS emulation | GCP services, local endpoints, console, and permitted non-production workflows |
| Production validation | Validate against AWS before production | Validate against Google Cloud before production |
If your application calls AWS APIs, use LocalStack. LocalCloud cannot serve S3, DynamoDB, Lambda, or SQS calls, and no amount of configuration will change that. LocalStack is also the older and more widely deployed project. LocalCloud is the right pick only when the code under test targets Google Cloud — BigQuery, Pub/Sub, Spanner, Bigtable, or Cloud Storage.
Teams building on both clouds commonly run both. The two runtimes bind different local ports and set different SDK environment variables, so they coexist in one development environment.
Start one Docker runtime, inspect its local services in the console, and work without a shared cloud development project for every iteration.
Browse the service catalogPoint local development and test clients at emulator endpoints, then remove the local configuration when exercising real Google Cloud.
See SDK examplesService behavior varies. Review supported capabilities and documented gaps before deciding which tests can stay local.
Review compatibilityYes. LocalCloud is a local Google Cloud runtime for non-production development, testing, CI, evaluation, and internal pilots permitted by its proprietary Public Preview License. It runs major GCP services in Docker so applications can use local endpoints instead of real cloud services.
No. LocalCloud is an independent product focused on Google Cloud workflows. LocalStack is an AWS-focused local cloud development platform.
No. LocalCloud is for permitted non-production workflows. Validate behavior against real Google Cloud before production deployment.
LocalCloud is designed for standard Google Cloud SDK and API workflows. Configure local emulator endpoints for development, then remove those local settings when validating against real Google Cloud.
Use LocalStack whenever the code under test calls AWS APIs. LocalCloud does not emulate AWS and is not a substitute for it. LocalStack is also the more established project, with a longer public track record and a wider published integration surface.
Yes. LocalCloud gives agent-written Google Cloud code a localhost target with no credentials or billing project by default, so an AI coding agent cannot create real cloud resources by mistake. It closes the dependency boundary; pair it with a code sandbox if you also need to contain arbitrary code execution.
The proprietary Public Preview License permits individuals and organizations, including for-profit companies, to use LocalCloud free of charge for internal development, testing, CI, evaluation, and internal pilots. Review the licensing documentation and compatibility details before adopting it for a team workflow.
Search text with 3+ characters is sent to PostHog. Do not enter secrets or personal data. Privacy details.
Type to search across all documentation pages
Try: "bigquery", "configuration", "seed data"