Category: RAG

• 1: vector
• 2: vchord
• 3: vectorscale
• 4: vectorize
• 5: pg_similarity
• 6: smlar
• 7: pg_summarize
• 8: pg_tiktoken
• 9: pg4ml
• 10: pgml

1 - vector

Overview

PGDG 1st Party Extension: pgvector : vector data type and ivfflat and hnsw access methods

Information

• Extension ID: 1800
• Extension Name: vector
• Package Name: pgvector
• Category: RAG
• License: PostgreSQL
• Website: https://github.com/pgvector/pgvector
• Language: C
• Extra Tags: N/A
• Comment: N/A

Metadata

• Latest Version: 0.8.0
• Postgres Support: 17,16,15,14,13
• Need Load: Shared library do not need explicit loading
• Need DDL: Need CREATE EXTENSION DDL
• Relocatable: Can not install to arbitrary schema
• Trusted: Untrusted, Require Superuser to Create
• Schemas: N/A
• Requires: N/A

RPM / DEB

• RPM Repo: PGDG
• RPM Name: pgvector_$v*
• RPM Ver : 0.8.0
• RPM Deps: N/A
• DEB Repo: PGDG
• DEB Name: postgresql-$v-pgvector
• DEB Ver : 0.8.0
• DEB Deps: N/A

Packages

Installation

Install vector via the pig CLI tool:

pig ext install pgvector; # Extension Namepig ext install vector; # normalized package name

Install pgvector via Pigsty playbook:

./pgsql.yml -t pg_extension -e '{"pg_extensions": ["pgvector"]}' # -l <cls>

Install pgvector RPM from YUM repo directly:

dnf install pgvector_17*;
dnf install pgvector_16*;
dnf install pgvector_15*;
dnf install pgvector_14*;
dnf install pgvector_13*;

Install pgvector DEB from APT repo directly:

apt install postgresql-17-pgvector;
apt install postgresql-16-pgvector;
apt install postgresql-15-pgvector;
apt install postgresql-14-pgvector;
apt install postgresql-13-pgvector;

Enable vector extension on PostgreSQL cluster：

CREATE EXTENSION vector;

2 - vchord

Overview

PIGSTY 3rd Party Extension: vchord : Vector database plugin for Postgres, written in Rust

Information

• Extension ID: 1810
• Extension Name: vchord
• Package Name: vchord
• Category: RAG
• License: AGPLv3
• Website: https://github.com/tensorchord/VectorChord
• Language: Rust
• Extra Tags: pgrx
• Comment:

Metadata

• Latest Version: 0.2.0
• Postgres Support: 17,16,15,14
• Need Load: Explicit Loading Required
• Need DDL: Need CREATE EXTENSION DDL
• Relocatable: Can not install to arbitrary schema
• Trusted: Untrusted, Require Superuser to Create
• Schemas: N/A
• Requires: vector

RPM / DEB

• RPM Repo: PIGSTY
• RPM Name: vchord_$v
• RPM Ver : 0.2.0
• RPM Deps: pgvector_$v
• DEB Repo: PIGSTY
• DEB Name: postgresql-$v-vchord
• DEB Ver : 0.1.0
• DEB Deps: postgresql-$v-pgvector

Packages

Installation

Install vchord via the pig CLI tool:

pig ext install vchord

Install vchord via Pigsty playbook:

./pgsql.yml -t pg_extension -e '{"pg_extensions": ["vchord"]}' # -l <cls>

Install vchord RPM from YUM repo directly:

dnf install vchord_17;
dnf install vchord_16;
dnf install vchord_15;
dnf install vchord_14;

Install vchord DEB from APT repo directly:

apt install postgresql-17-vchord;
apt install postgresql-16-vchord;
apt install postgresql-15-vchord;
apt install postgresql-14-vchord;

Extension vchord has to be added to shared_preload_libraries

shared_preload_libraries = 'vchord'; # add to pg cluster config

Enable vchord extension on PostgreSQL cluster：

CREATE EXTENSION vchord CASCADE;

Usage

• https://github.com/tensorchord/VectorChord
• Launch Blog: VectorChord: Store 400k Vectors for $1 in PostgreSQL

Add this extension to shared_preload_libraries in postgresql.conf

CREATE EXTENSION vchord CASCADE;

Create Index on embedding:

CREATE INDEX ON gist_train USING vchordrq (embedding vector_l2_ops) WITH (options = $$
residual_quantization = true
[build.internal]
lists = [4096]
spherical_centroids = false
$$);

Docs

Query

The query statement is exactly the same as pgvector. VectorChord supports any filter operation and WHERE/JOIN clauses like pgvecto.rs with VBASE.

SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;

Supported distance functions are:

• <-> - L2 distance
• <#> - (negative) inner product
• <=> - cosine distance

Query Performance Tuning

You can fine-tune the search performance by adjusting the probes and epsilon parameters:

-- Set probes to control the number of lists scanned. 
-- Recommended range: 3%–10% of the total `lists` value.
SET vchordrq.probes = 100;

-- Set epsilon to control the reranking precision.
-- Larger value means more rerank for higher recall rate.
-- Don't change it unless you only have limited memory.
-- Recommended range: 1.0–1.9. Default value is 1.9.
SET vchordrq.epsilon = 1.9;

-- vchordrq relies on a projection matrix to optimize performance.
-- Add your vector dimensions to the `prewarm_dim` list to reduce latency.
-- If this is not configured, the first query will have higher latency as the matrix is generated on demand.
-- Default value: '64,128,256,384,512,768,1024,1536'
-- Note: This setting requires a database restart to take effect.
ALTER SYSTEM SET vchordrq.prewarm_dim = '64,128,256,384,512,768,1024,1536';

And for postgres’s setting

-- If using SSDs, set `effective_io_concurrency` to 200 for faster disk I/O.
SET effective_io_concurrency = 200;

-- Disable JIT (Just-In-Time Compilation) as it offers minimal benefit (1–2%) 
-- and adds overhead for single-query workloads.
SET jit = off;

-- Allocate at least 25% of total memory to `shared_buffers`. 
-- For disk-heavy workloads, you can increase this to up to 90% of total memory. You may also want to disable swap with network storage to avoid io hang.
-- Note: A restart is required for this setting to take effect.
ALTER SYSTEM SET shared_buffers = '8GB';

Indexing prewarm

To prewarm the index, you can use the following SQL. It will significantly improve performance when using limited memory.

-- vchordrq_prewarm(index_name::regclass) to prewarm the index into the shared buffer
SELECT vchordrq_prewarm('gist_train_embedding_idx'::regclass)"

Index Build Time

Index building can parallelized, and with external centroid precomputation, the total time is primarily limited by disk speed. Optimize parallelism using the following settings:

-- Set this to the number of CPU cores available for parallel operations.
SET max_parallel_maintenance_workers = 8;
SET max_parallel_workers = 8;

-- Adjust the total number of worker processes. 
-- Note: A restart is required for this setting to take effect.
ALTER SYSTEM SET max_worker_processes = 8;

Indexing Progress

You can check the indexing progress by querying the pg_stat_progress_create_index view.

SELECT phase, round(100.0 * blocks_done / nullif(blocks_total, 0), 1) AS "%" FROM pg_stat_progress_create_index;

External Index Precomputation

Unlike pure SQL, an external index precomputation will first do clustering outside and insert centroids to a PostgreSQL table. Although it might be more complicated, external build is definitely much faster on larger dataset (>5M).

To get started, you need to do a clustering of vectors using faiss, scikit-learn or any other clustering library.

The centroids should be preset in a table of any name with 3 columns:

• id(integer): id of each centroid, should be unique
• parent(integer, nullable): parent id of each centroid, should be NULL for normal clustering
• vector(vector): representation of each centroid, pgvector vector type

And example could be like this:

-- Create table of centroids
CREATE TABLE public.centroids (id integer NOT NULL UNIQUE, parent integer, vector vector(768));
-- Insert centroids into it
INSERT INTO public.centroids (id, parent, vector) VALUES (1, NULL, '{0.1, 0.2, 0.3, ..., 0.768}');
INSERT INTO public.centroids (id, parent, vector) VALUES (2, NULL, '{0.4, 0.5, 0.6, ..., 0.768}');
INSERT INTO public.centroids (id, parent, vector) VALUES (3, NULL, '{0.7, 0.8, 0.9, ..., 0.768}');
-- ...

-- Create index using the centroid table
CREATE INDEX ON gist_train USING vchordrq (embedding vector_l2_ops) WITH (options = $$
[build.external]
table = 'public.centroids'
$$);

To simplify the workflow, we provide end-to-end scripts for external index pre-computation, see scripts.

Limitations

• Data Type Support: Currently, only the f32 data type is supported for vectors.
• Architecture Compatibility: The fast-scan kernel is optimized for x86_64 architectures. While it runs on aarch64, performance may be lower.
• KMeans Clustering: The built-in KMeans clustering is not yet fully optimized and may require substantial memory. We strongly recommend using external centroid precomputation for efficient index construction.

3 - vectorscale

Overview

PIGSTY 3rd Party Extension: pgvectorscale : pgvectorscale: Advanced indexing for vector data

Information

• Extension ID: 1820
• Extension Name: vectorscale
• Package Name: pgvectorscale
• Category: RAG
• License: PostgreSQL
• Website: https://github.com/timescale/pgvectorscale
• Language: Rust
• Extra Tags: pgrx
• Comment: N/A

Metadata

• Latest Version: 0.5.1
• Postgres Support: 17,16,15,14,13
• Need Load: Shared library do not need explicit loading
• Need DDL: Need CREATE EXTENSION DDL
• Relocatable: Can be installed into other schemas
• Trusted: Untrusted, Require Superuser to Create
• Schemas: N/A
• Requires: vector

RPM / DEB

• RPM Repo: PIGSTY
• RPM Name: pgvectorscale_$v
• RPM Ver : 0.5.1
• RPM Deps: N/A
• DEB Repo: PIGSTY
• DEB Name: postgresql-$v-pgvectorscale
• DEB Ver : 0.5.1
• DEB Deps: N/A

Packages

Installation

Install vectorscale via the pig CLI tool:

pig ext install pgvectorscale; # Extension Namepig ext install vectorscale; # normalized package name

Install pgvectorscale via Pigsty playbook:

./pgsql.yml -t pg_extension -e '{"pg_extensions": ["pgvectorscale"]}' # -l <cls>

Install pgvectorscale RPM from YUM repo directly:

dnf install pgvectorscale_17;
dnf install pgvectorscale_16;
dnf install pgvectorscale_15;
dnf install pgvectorscale_14;
dnf install pgvectorscale_13;

Install pgvectorscale DEB from APT repo directly:

apt install postgresql-17-pgvectorscale;
apt install postgresql-16-pgvectorscale;
apt install postgresql-15-pgvectorscale;
apt install postgresql-14-pgvectorscale;
apt install postgresql-13-pgvectorscale;

Enable vectorscale extension on PostgreSQL cluster：

CREATE EXTENSION vectorscale CASCADE;

4 - vectorize

Overview

PIGSTY 3rd Party Extension: pg_vectorize : The simplest way to do vector search on Postgres

Information

• Extension ID: 1830
• Extension Name: vectorize
• Package Name: pg_vectorize
• Category: RAG
• License: PostgreSQL
• Website: https://github.com/tembo-io/pg_vectorize
• Language: Rust
• Extra Tags: pgrx
• Comment: N/A

Metadata

• Latest Version: 0.21.1
• Postgres Support: 17,16,15,14
• Need Load: Shared library do not need explicit loading
• Need DDL: Need CREATE EXTENSION DDL
• Relocatable: Can be installed into other schemas
• Trusted: Untrusted, Require Superuser to Create
• Schemas: vectorize
• Requires: pg_cron, pgmq, vector

RPM / DEB

• RPM Repo: PIGSTY
• RPM Name: pg_vectorize_$v
• RPM Ver : 0.21.1
• RPM Deps: pgmq_$v, pg_cron_$v, pgvector_$v
• DEB Repo: PIGSTY
• DEB Name: postgresql-$v-pg-vectorize
• DEB Ver : 0.21.1
• DEB Deps: postgresql-$v-pgmq, postgresql-$v-pg-cron, postgresql-$v-pgvector

Packages

Installation

Install vectorize via the pig CLI tool:

pig ext install pg_vectorize; # Extension Namepig ext install vectorize; # normalized package name

Install pg_vectorize via Pigsty playbook:

./pgsql.yml -t pg_extension -e '{"pg_extensions": ["pg_vectorize"]}' # -l <cls>

Install pg_vectorize RPM from YUM repo directly:

dnf install pg_vectorize_17;
dnf install pg_vectorize_16;
dnf install pg_vectorize_15;
dnf install pg_vectorize_14;

Install pg_vectorize DEB from APT repo directly:

apt install postgresql-17-pg-vectorize;
apt install postgresql-16-pg-vectorize;
apt install postgresql-15-pg-vectorize;
apt install postgresql-14-pg-vectorize;

Enable vectorize extension on PostgreSQL cluster：

CREATE EXTENSION vectorize CASCADE;

5 - pg_similarity

Overview

MIXED 3rd Party Extension: pg_similarity : support similarity queries

Information

• Extension ID: 1840
• Extension Name: pg_similarity
• Package Name: pg_similarity
• Category: RAG
• License: BSD 3
• Website: https://github.com/eulerto/pg_similarity
• Language: C
• Extra Tags: N/A
• Comment: N/A

Metadata

• Latest Version: 1.0
• Postgres Support: 17,16,15,14,13
• Need Load: Shared library do not need explicit loading
• Need DDL: Need CREATE EXTENSION DDL
• Relocatable: Can not install to arbitrary schema
• Trusted: Untrusted, Require Superuser to Create
• Schemas: N/A
• Requires: N/A

RPM / DEB

• RPM Repo: PIGSTY
• RPM Name: pg_similarity_$v*
• RPM Ver : 1.0
• RPM Deps: N/A
• DEB Repo: PGDG
• DEB Name: postgresql-$v-similarity
• DEB Ver : 1.0
• DEB Deps: N/A

Packages

Installation

Install pg_similarity via the pig CLI tool:

pig ext install pg_similarity

Install pg_similarity via Pigsty playbook:

./pgsql.yml -t pg_extension -e '{"pg_extensions": ["pg_similarity"]}' # -l <cls>

Install pg_similarity RPM from YUM repo directly:

dnf install pg_similarity_17*;
dnf install pg_similarity_16*;
dnf install pg_similarity_15*;
dnf install pg_similarity_14*;
dnf install pg_similarity_13*;

Install pg_similarity DEB from APT repo directly:

apt install postgresql-17-similarity;
apt install postgresql-16-similarity;
apt install postgresql-15-similarity;
apt install postgresql-14-similarity;
apt install postgresql-13-similarity;

Enable pg_similarity extension on PostgreSQL cluster：

CREATE EXTENSION pg_similarity;

6 - smlar

Overview

PIGSTY 3rd Party Extension: smlar : Effective similarity search

Information

• Extension ID: 1850
• Extension Name: smlar
• Package Name: smlar
• Category: RAG
• License: PostgreSQL
• Website: https://github.com/jirutka/smlar
• Language: C
• Extra Tags: nil-lic
• Comment:

Metadata

• Latest Version: 1.0
• Postgres Support: 17,16,15,14,13
• Need Load: Shared library do not need explicit loading
• Need DDL: Need CREATE EXTENSION DDL
• Relocatable: Can not install to arbitrary schema
• Trusted: Unknown
• Schemas: N/A
• Requires: N/A

RPM / DEB

• RPM Repo: PIGSTY
• RPM Name: smlar_$v*
• RPM Ver : 1.0
• RPM Deps: N/A
• DEB Repo: PIGSTY
• DEB Name: postgresql-$v-smlar
• DEB Ver : 1.0
• DEB Deps: N/A

Packages

Installation

Install smlar via the pig CLI tool:

pig ext install smlar

Install smlar via Pigsty playbook:

./pgsql.yml -t pg_extension -e '{"pg_extensions": ["smlar"]}' # -l <cls>

Install smlar RPM from YUM repo directly:

dnf install smlar_17*;
dnf install smlar_16*;
dnf install smlar_15*;
dnf install smlar_14*;
dnf install smlar_13*;

Install smlar DEB from APT repo directly:

apt install postgresql-17-smlar;
apt install postgresql-16-smlar;
apt install postgresql-15-smlar;
apt install postgresql-14-smlar;
apt install postgresql-13-smlar;

Enable smlar extension on PostgreSQL cluster：

CREATE EXTENSION smlar;

7 - pg_summarize

Overview

PIGSTY 3rd Party Extension: pg_summarize : Text Summarization using LLMs. Built using pgrx

Information

• Extension ID: 1860
• Extension Name: pg_summarize
• Package Name: pg_summarize
• Category: RAG
• License: PostgreSQL
• Website: https://github.com/HexaCluster/pg_summarize
• Language: Rust
• Extra Tags: pgrx
• Comment: N/A

Metadata

• Latest Version: 0.0.1
• Postgres Support: 17,16,15,14,13
• Need Load: Shared library do not need explicit loading
• Need DDL: Need CREATE EXTENSION DDL
• Relocatable: Can be installed into other schemas
• Trusted: Untrusted, Require Superuser to Create
• Schemas: N/A
• Requires: N/A

RPM / DEB

• RPM Repo: PIGSTY
• RPM Name: pg_summarize_$v
• RPM Ver : 0.0.1
• RPM Deps: N/A
• DEB Repo: PIGSTY
• DEB Name: postgresql-$v-pg-summarize
• DEB Ver : 0.0.1
• DEB Deps: N/A

Packages

Installation

Install pg_summarize via the pig CLI tool:

pig ext install pg_summarize

Install pg_summarize via Pigsty playbook:

./pgsql.yml -t pg_extension -e '{"pg_extensions": ["pg_summarize"]}' # -l <cls>

Install pg_summarize RPM from YUM repo directly:

dnf install pg_summarize_17;
dnf install pg_summarize_16;
dnf install pg_summarize_15;
dnf install pg_summarize_14;
dnf install pg_summarize_13;

Install pg_summarize DEB from APT repo directly:

apt install postgresql-17-pg-summarize;
apt install postgresql-16-pg-summarize;
apt install postgresql-15-pg-summarize;
apt install postgresql-14-pg-summarize;
apt install postgresql-13-pg-summarize;

Enable pg_summarize extension on PostgreSQL cluster：

CREATE EXTENSION pg_summarize;

8 - pg_tiktoken

Overview

PIGSTY 3rd Party Extension: pg_tiktoken : pg_tictoken: tiktoken tokenizer for use with OpenAI models in postgres

Information

• Extension ID: 1870
• Extension Name: pg_tiktoken
• Package Name: pg_tiktoken
• Category: RAG
• License: Apache-2.0
• Website: https://github.com/kelvich/pg_tiktoken
• Language: Rust
• Extra Tags: pgrx
• Comment: N/A

Metadata

• Latest Version: 0.0.1
• Postgres Support: 17,16,15,14,13
• Need Load: Shared library do not need explicit loading
• Need DDL: Need CREATE EXTENSION DDL
• Relocatable: Can be installed into other schemas
• Trusted: Untrusted, Require Superuser to Create
• Schemas: N/A
• Requires: N/A

RPM / DEB

• RPM Repo: PIGSTY
• RPM Name: pg_tiktoken_$v
• RPM Ver : 0.0.1
• RPM Deps: N/A
• DEB Repo: PIGSTY
• DEB Name: postgresql-$v-pg-tiktoken
• DEB Ver : 0.0.1
• DEB Deps: N/A

Packages

Installation

Install pg_tiktoken via the pig CLI tool:

pig ext install pg_tiktoken

Install pg_tiktoken via Pigsty playbook:

./pgsql.yml -t pg_extension -e '{"pg_extensions": ["pg_tiktoken"]}' # -l <cls>

Install pg_tiktoken RPM from YUM repo directly:

dnf install pg_tiktoken_17;
dnf install pg_tiktoken_16;
dnf install pg_tiktoken_15;
dnf install pg_tiktoken_14;
dnf install pg_tiktoken_13;

Install pg_tiktoken DEB from APT repo directly:

apt install postgresql-17-pg-tiktoken;
apt install postgresql-16-pg-tiktoken;
apt install postgresql-15-pg-tiktoken;
apt install postgresql-14-pg-tiktoken;
apt install postgresql-13-pg-tiktoken;

Enable pg_tiktoken extension on PostgreSQL cluster：

CREATE EXTENSION pg_tiktoken;

9 - pg4ml

Overview

PIGSTY 3rd Party Extension: pg4ml : Machine learning framework for PostgreSQL

Information

• Extension ID: 1880
• Extension Name: pg4ml
• Package Name: pg4ml
• Category: RAG
• License: AGPLv3
• Website: https://gitee.com/guotiecheng/plpgsql_pg4ml
• Language: C
• Extra Tags: N/A
• Comment:

Metadata

• Latest Version: 2.0
• Postgres Support: 17,16,15,14,13
• Need Load: Shared library do not need explicit loading
• Need DDL: Need CREATE EXTENSION DDL
• Relocatable: Can not install to arbitrary schema
• Trusted: Trusted, Can be created by user with CREATE Privilege
• Schemas: N/A
• Requires: plpgsql, tablefunc, cube, plpython3u

RPM / DEB

• RPM Repo: PIGSTY
• RPM Name: pg4ml_$v
• RPM Ver : 2.0
• RPM Deps: N/A
• DEB Repo: PIGSTY
• DEB Name: postgresql-$v-pg4ml
• DEB Ver : 2.0
• DEB Deps: N/A

Packages

Installation

Install pg4ml via the pig CLI tool:

pig ext install pg4ml

Install pg4ml via Pigsty playbook:

./pgsql.yml -t pg_extension -e '{"pg_extensions": ["pg4ml"]}' # -l <cls>

Install pg4ml RPM from YUM repo directly:

dnf install pg4ml_17;
dnf install pg4ml_16;
dnf install pg4ml_15;
dnf install pg4ml_14;
dnf install pg4ml_13;

Install pg4ml DEB from APT repo directly:

apt install postgresql-17-pg4ml;
apt install postgresql-16-pg4ml;
apt install postgresql-15-pg4ml;
apt install postgresql-14-pg4ml;
apt install postgresql-13-pg4ml;

Enable pg4ml extension on PostgreSQL cluster：

CREATE EXTENSION pg4ml CASCADE;

10 - pgml

Overview

PIGSTY 3rd Party Extension: pgml : PostgresML: Run AL/ML workloads with SQL interface

Information

• Extension ID: 1890
• Extension Name: pgml
• Package Name: pgml
• Category: RAG
• License: MIT
• Website: https://github.com/postgresml/postgresml
• Language: Rust
• Extra Tags: pgrx
• Comment: N/A

Metadata

• Latest Version: 2.10.0
• Postgres Support: 17,16,15,14
• Need Load: Explicit Loading Required
• Need DDL: Need CREATE EXTENSION DDL
• Relocatable: Can be installed into other schemas
• Trusted: Untrusted, Require Superuser to Create
• Schemas: pgml
• Requires: N/A

RPM / DEB

• RPM Repo: None
• RPM Name: N/A
• RPM Ver : N/A
• RPM Deps: N/A
• DEB Repo: PIGSTY
• DEB Name: postgresql-$v-pgml
• DEB Ver : 2.10.0
• DEB Deps: N/A

Packages

Installation

Install pgml via the pig CLI tool:

pig ext install pgml

Install pgml via Pigsty playbook:

./pgsql.yml -t pg_extension -e '{"pg_extensions": ["pgml"]}' # -l <cls>

Install pgml DEB from APT repo directly:

apt install postgresql-17-pgml;
apt install postgresql-16-pgml;
apt install postgresql-15-pgml;
apt install postgresql-14-pgml;

Extension pgml has to be added to shared_preload_libraries

shared_preload_libraries = 'pgml'; # add to pg cluster config

Enable pgml extension on PostgreSQL cluster：

CREATE EXTENSION pgml;

Usage

After installing the pgml extension and python dependencies on all cluster nodes, you can enable pgml on the PostgreSQL cluster.

Configure cluster with patronictl command and add pgml to shared_preload_libraries, and specify your venv dir in pgml.venv:

shared_preload_libraries: pgml, timescaledb, pg_stat_statements, auto_explain
pgml.venv: '/data/pgml'

After that, restart database cluster, and create extension with SQL command:

CREATE EXTENSION vector;        -- nice to have pgvector installed too!
CREATE EXTENSION pgml;          -- create PostgresML in current database
SELECT pgml.version();          -- print PostgresML version string

If it works, you should see something like:

# create extension pgml;
INFO:  Python version: 3.11.2 (main, Oct  5 2023, 16:06:03) [GCC 8.5.0 20210514 (Red Hat 8.5.0-18)]
INFO:  Scikit-learn 1.3.0, XGBoost 2.0.0, LightGBM 4.1.0, NumPy 1.26.1
CREATE EXTENSION

# SELECT pgml.version(); -- print PostgresML version string
 version
---------
 2.7.8

You are all set! Check PostgresML for more details: https://postgresml.org/docs/guides/use-cases/