About pgrust
pgrust is a rewrite of PostgreSQL in Rust. Its authors describe it as an experiment in what Postgres might look like if it were designed today, and as a project built with heavy use of AI-assisted development. It targets drop-in compatibility with the Postgres wire protocol and SQL dialect.
The README claims the project passes all 46,066 tests in the Postgres regression suite, and states that it is faster than both Postgres and ClickHouse; those performance claims come from the project itself and should be verified for your workload. A browser demo, a Discord community and a talk about JIT compilers and AI-driven development are linked from the repository. The current release is v0.3, so it is early-stage software.
pgrust is licensed under AGPL-3.0 and is intended to be run on your own machines. It is of interest to database engineers, researchers and Postgres users curious about a memory-safe, Rust-based engine, and it is not yet a proven replacement for production Postgres.
Key features
- Rust reimplementation of PostgreSQL
- Wire protocol compatible with Postgres
- SQL dialect compatibility
- Runs the Postgres regression test suite
- Online browser demo available
- Early-stage v0.3 release
Good fit for
- →Experimenting with a Rust Postgres engine
- →Benchmarking against Postgres workloads
- Tags
- postgres
- rust
- database
- sql
- experimental
- postgresql-compatible
- ai-assisted
pgrust: questions and answers
- What is pgrust used for?
- pgrust is an experimental Rust reimplementation of PostgreSQL that aims to stay wire and SQL compatible while outperforming Postgres and ClickHouse. It is a good fit for experimenting with a Rust Postgres engine and benchmarking against Postgres workloads.
- Is pgrust open source?
- Yes. pgrust is open source under the AGPL-3.0 licence. Its source code is on GitHub at malisper/pgrust and is written mainly in Rust.
- Is pgrust free?
- Yes. pgrust is open source, so the software itself is free to use.
- Can I self-host pgrust?
- Yes. pgrust can be self-hosted on your own server or infrastructure.
- What are some alternatives to pgrust?
- Similar open-source tools in the Databases category include SurrealDB, YugabyteDB and Materialize. SaaS products in the same category include Cloudflare D1, Aerospike and Amazon Athena.
- Is pgrust actively maintained?
- Yes. The most recent commit to pgrust was on 18 September 2026, and the latest release is v0.3, published on 15 September 2026. The project has 5.2k stars on GitHub.
Open-source alternatives to pgrust
See all
SurrealDB
Databases
A scalable, distributed, collaborative, document-graph database, for the realtime web
OSSvs MongoDB Atlas★ 33k
YugabyteDB
Databases
YugabyteDB - the cloud native distributed SQL database for mission-critical applications.
OSSvs Google Cloud Spanner★ 11k
Materialize
Databases
The live data layer for apps and AI agents. Create up-to-the-second views into your busine
OSS★ 6.4k
GlareDB
Databases
GlareDB: A light and fast SQL database for analytics
MITvs MotherDuck★ 1k
ClickHouse
Databases
ClickHouse is a fast open-source column-oriented database management system that allows ge
Apache-2.0vs Snowflake★ 50k
DuckDB
Databases
DuckDB is an analytical in-process SQL database management system
MITvs MotherDuck★ 42k
SaaS alternatives to pgrust
See all
Cloudflare D1
Databases
Serverless SQLite-based database for Cloudflare Workers
SaaS
Aerospike
Databases
Real-time NoSQL database for low-latency, high-throughput applications
SaaS
Amazon Athena
Databases
Serverless interactive query service for analyzing data in Amazon S3 with SQL
SaaS
Amazon MemoryDB
Databases
Durable in-memory database service on AWS compatible with Valkey and Redis OSS
SaaS
Amazon Redshift
Databases
Managed cloud data warehouse on AWS for SQL analytics at scale
SaaS
Azure Data Explorer
Databases
Analytics service on Azure for exploring log, telemetry and time-series data with Kusto queries
SaaS

