Halt LogoHalt
HomeDocsExamplesChangelogBlog
GitHub

Getting Started

IntroductionInstallationQuick StartInteractive SetupNew

Core Concepts

AlgorithmsPoliciesKey StrategiesDynamic LimitsNew

Storage Backends

In-MemoryRedisResilienceNewPostgreSQLMongoDBDynamoDBMemcached

Framework Tutorials

FastAPIFlaskDjangoExpressNext.jsHonoFastifyGraphQL (Apollo)

Runtime

Edge & Runtime Support

SaaS Features

Plan-Based LimitingQuota ManagementPenalty SystemTelemetry

Observability

OverviewNew

Storage Backends

Halt keeps the same algorithms and policies across every backend pick the one that fits your infrastructure. For anything running more than one process, use Redis.

Redis

Recommended for production

Atomic (Lua), cluster-safe, fail-open, sync + async.

In-Memory

Development

Zero-setup store for development and single-process apps.

PostgreSQL

Durable, relational storage shared across instances.

MongoDB

Document storage with TTL indexes.

DynamoDB

AWS serverless, auto-scaling key-value storage.

Memcached

Low-latency distributed cache.

Halt Logo

Modern Rate Limiting

Halt

SaaS-aware, cross-language rate limiting — users, API keys, plans, quotas, weighted endpoints, abuse controls, atomic Redis, and built-in observability.

Documentation

OverviewInstallationAlgorithmsExamples

Resources

ChangelogBlogGitHubContributingReport an issue

Get the package

halt-rate on npmhalt-rate on PyPI
npm versionPyPI version
GitHubTwitter

© 2026 Halt. MIT Licensed.