Hasura is an open-source GraphQL engine that connects to a database and instantly generates a complete GraphQL and REST API with built-in permissions, real-time subscriptions and webhook triggers. It was founded in 2017 by Tanmai Gopal and Rajoshi Ghosh and is headquartered in San Francisco with a major engineering presence in Bangalore, India.
Instead of resolving GraphQL queries field by field, Hasura compiles incoming queries directly into a single optimized SQL statement, avoiding the N+1 query problem common in hand-written GraphQL resolvers and giving Hasura a reputation for very low latency and low resource usage.
Key Features
Hasura automatically generates CRUD operations, real-time subscriptions and fine-grained row and column-level permissions from a database schema, with permission rules able to reference dynamic variables from an organization's existing authentication system. Event triggers fire webhooks on database changes, and scheduled or cron triggers support background jobs.
Actions let teams wire custom business logic written in any language into the unified GraphQL schema, while Remote Schemas and Remote Joins federate multiple GraphQL APIs and external data sources into one endpoint. Beyond PostgreSQL, Hasura's Data Connector framework extends support to MySQL, SQL Server, Oracle, MongoDB, Snowflake and BigQuery.
Pricing
The open-source Hasura GraphQL Engine is free to self-host with no licensing cost. Hasura Cloud's Free tier requires no credit card and supports up to 3 million API requests per month across up to three projects.
The Cloud Professional tier uses pay-as-you-go pricing at 1.50 dollars per active-hour for Postgres projects and 3.00 dollars per active-hour for advanced database connectors, plus 0.13 dollars per gigabyte of data passthrough. Cloud Enterprise offers unlimited requests, dedicated infrastructure and 24/7 support under a custom annual contract requiring a minimum 12-month upfront commitment.
Key Features
Instant GraphQL and REST APIs — Introspects a connected database schema and automatically generates a complete GraphQL and REST API with no manual resolver code.
Compiler-Based Query Execution — Compiles nested GraphQL queries into a single optimized SQL statement, avoiding the N+1 query problem common in hand-written resolvers.
Fine-Grained Permissions — Row-level and column-level access rules that can incorporate dynamic variables from an existing authentication system, such as user role or tenant ID.
Real-Time Subscriptions — Pushes live updates to connected clients automatically as underlying database data changes, without additional infrastructure.
Event and Scheduled Triggers — Fires webhooks on database insert, update or delete events, and supports cron-based scheduled triggers for background jobs.
Actions for Custom Business Logic — Wires custom logic written in any programming language into the unified GraphQL schema alongside auto-generated operations.
Remote Schemas and Joins — Federates multiple GraphQL APIs and external data sources into a single unified endpoint.
Multi-Database Connectors — Extends beyond PostgreSQL to MySQL, SQL Server, Oracle, MongoDB, Snowflake and BigQuery through the Data Connector framework.
Pros & Cons
Pros
Free, full-featured open-source core that can be self-hosted with no licensing cost
Compiler-based architecture avoids the N+1 query problem and delivers strong performance with low resource consumption
Fine-grained, dynamic permissions system is more mature than many newer instant-API competitors
Real-time subscriptions and event triggers are built in without needing separate infrastructure
Supports multiple databases beyond Postgres, including MongoDB, Snowflake and BigQuery
Cons
Cloud Professional's usage-based pricing (active-hours plus data passthrough) can be harder to predict than flat per-seat pricing
Cloud Enterprise requires a minimum 12-month upfront commitment, which is a barrier for teams wanting to test at scale first
Advanced database connectors cost twice as much per active-hour as the Postgres connector on Cloud Professional
Teams with highly custom business logic may still need to write Actions handlers, reducing some of the no-code appeal
Smaller company size relative to larger backend-as-a-service platforms may mean a narrower support and integration ecosystem
Pricing
Free $0 N/A, no credit card required
Cloud Professional From $1.50/active-hour (Postgres) or $3.00/active-hour (advanced connectors) Monthly, pay-as-you-go
Hasura is used to instantly generate a GraphQL and REST API from a database, with built-in permissions, real-time subscriptions and webhook triggers, removing the need to hand-write backend resolver code.
Is Hasura free to use
Yes, the open-source Hasura GraphQL Engine is free to self-host. Hasura Cloud also has a free tier supporting up to 3 million API requests per month with no credit card required.
Does Hasura only work with PostgreSQL
No, while PostgreSQL was Hasura's original focus, its Data Connector framework now supports MySQL, SQL Server, Oracle, MongoDB, Snowflake and BigQuery, among other data sources.
Who founded Hasura
Hasura was founded in 2017 by Tanmai Gopal and Rajoshi Ghosh, and the company is headquartered in San Francisco with a major engineering team in Bangalore, India.
How does Hasura avoid the GraphQL N+1 problem
Hasura compiles an incoming GraphQL query directly into a single optimized SQL statement rather than resolving fields one at a time, avoiding the multiple round-trip database queries that cause the N+1 problem in hand-written resolvers.
How much funding has Hasura raised
Hasura has raised approximately 239 million dollars in total funding, including a 100 million dollar Series C round in February 2022 that valued the company at approximately 1 billion dollars.