Open-source vector similarity search for Postgres
Supports exact and approximate nearest neighbor search for L2 distance, inner product, and cosine distance
Compile and install the extension (supports Postgres 11+)
cd /tmp
git clone --branch v0.4.1 https://github.com/pgvector/pgvector.git
cd pgvector
make
make install # may need sudo
Then load it in databases where you want to use it
CREATE EXTENSION vector;
You can also install it with Docker, Homebrew, PGXN, Yum, or conda-forge
Create a vector column with 3 dimensions
CREATE TABLE items (id bigserial PRIMARY KEY, embedding vector(3));
Insert values
INSERT INTO items (embedding) VALUES ('[1,2,3]'), ('[4,5,6]');
Get the nearest neighbors by L2 distance
SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
Also supports inner product (<#>
) and cosine distance (<=>
)
Note: <#>
returns the negative inner product since Postgres only supports ASC
order index scans on operators
Insert vectors
INSERT INTO items (embedding) VALUES ('[1,2,3]'), ('[4,5,6]');
Upsert vectors
INSERT INTO items (id, embedding) VALUES (1, '[1,2,3]'), (2, '[4,5,6]')
ON CONFLICT (id) DO UPDATE SET embedding = EXCLUDED.embedding;
Update vectors
UPDATE items SET embedding = '[1,2,3]' WHERE id = 1;
Delete vectors
DELETE FROM items WHERE id = 1;
Use a SELECT
clause to get the distance
SELECT embedding <-> '[3,1,2]' AS distance FROM items;
For cosine similarity, use 1 - cosine distance
SELECT 1 - (embedding <=> '[3,1,2]') AS similarity FROM items;
Use a WHERE
clause to get rows within a certain distance
SELECT * FROM items WHERE embedding <-> '[3,1,2]' < 5;
Note: Combine with ORDER BY
and LIMIT
to use an index
Average vectors
SELECT AVG(embedding) FROM items;
Average groups of vectors
SELECT category_id, AVG(embedding) FROM items GROUP BY category_id;
By default, pgvector performs exact nearest neighbor search, which provides perfect recall.
You can add an index to use approximate nearest neighbor search, which trades some recall for performance. Unlike typical indexes, you will see different results for queries after adding an approximate index.
Two keys to achieving good recall are:
- Create the index after the table has some data
- Choose an appropriate number of lists (lower is better for recall, higher is better for speed)
A good place to start is:
rows / 1000
for up to 1M rowssqrt(rows)
for over 1M rows
Add an index for each distance function you want to use.
L2 distance
CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WITH (lists = 100);
Inner product
CREATE INDEX ON items USING ivfflat (embedding vector_ip_ops) WITH (lists = 100);
Cosine distance
CREATE INDEX ON items USING ivfflat (embedding vector_cosine_ops) WITH (lists = 100);
Vectors with up to 2,000 dimensions can be indexed.
Specify the number of probes (1 by default)
SET ivfflat.probes = 1;
A higher value provides better recall at the cost of speed, and it can be set to the number of lists for exact nearest neighbor search (at which point the planner won’t use the index)
Use SET LOCAL
inside a transaction to set it for a single query
BEGIN;
SET LOCAL ivfflat.probes = 1;
SELECT ...
COMMIT;
Check indexing progress with Postgres 12+
SELECT phase, tuples_done, tuples_total FROM pg_stat_progress_create_index;
The phases are:
initializing
performing k-means
sorting tuples
loading tuples
Note: tuples_done
and tuples_total
are only populated during the loading tuples
phase
Consider partial indexes for queries with a WHERE
clause
SELECT * FROM items WHERE category_id = 123 ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
can be indexed with:
CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WITH (lists = 100) WHERE (category_id = 123);
To index many different values of category_id
, consider partitioning on category_id
.
CREATE TABLE items (embedding vector(3), category_id int) PARTITION BY LIST(category_id);
Use EXPLAIN ANALYZE
to debug performance.
EXPLAIN ANALYZE SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 1;
To speed up queries without an index, increase max_parallel_workers_per_gather
.
SET max_parallel_workers_per_gather = 4;
If vectors are normalized to length 1 (like OpenAI embeddings), use inner product for best performance.
SELECT * FROM items ORDER BY embedding <#> '[3,1,2]' LIMIT 1;
To speed up queries with an index, increase the number of inverted lists (at the expense of recall).
CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WITH (lists = 1000);
Use pgvector from any language with a Postgres client. You can even generate and store vectors in one language and query them in another.
Language | Libraries / Examples |
---|---|
C++ | pgvector-cpp |
C# | pgvector-dotnet |
Elixir | pgvector-elixir |
Go | pgvector-go |
Java, Scala | pgvector-java |
Julia | pgvector-julia |
Lua | pgvector-lua |
Node.js | pgvector-node |
Perl | pgvector-perl |
PHP | pgvector-php |
Python | pgvector-python |
R | pgvector-r |
Ruby | pgvector-ruby, Neighbor |
Rust | pgvector-rust |
A non-partitioned table has a limit of 32 TB by default in Postgres. A partitioned table can have thousands of partitions of that size.
Yes, pgvector uses the write-ahead log (WAL), which allows for replication and point-in-time recovery.
Two things you can try are:
- use dimensionality reduction
- compile Postgres with a larger block size (
./configure --with-blocksize=32
) and edit the limit insrc/ivfflat.h
Each vector takes 4 * dimensions + 8
bytes of storage. Each element is a single precision floating-point number (like the real
type in Postgres), and all elements must be finite (no NaN
, Infinity
or -Infinity
). Vectors can have up to 16,000 dimensions.
Operator | Description |
---|---|
+ | element-wise addition |
- | element-wise subtraction |
<-> | Euclidean distance |
<#> | negative inner product |
<=> | cosine distance |
Function | Description |
---|---|
cosine_distance(vector, vector) → double precision | cosine distance |
inner_product(vector, vector) → double precision | inner product |
l2_distance(vector, vector) → double precision | Euclidean distance |
vector_dims(vector) → integer | number of dimensions |
vector_norm(vector) → double precision | Euclidean norm |
Function | Description |
---|---|
avg(vector) → vector | arithmetic mean |
Get the Docker image with:
docker pull ankane/pgvector
This adds pgvector to the Postgres image (run it the same way).
You can also build the image manually:
git clone --branch v0.4.1 https://github.com/pgvector/pgvector.git
cd pgvector
docker build -t pgvector .
With Homebrew Postgres, you can use:
brew install pgvector
Install from the PostgreSQL Extension Network with:
pgxn install vector
RPM packages are available from the PostgreSQL Yum Repository. Follow the setup instructions for your distribution and run:
sudo yum install pgvector_15
# or
sudo dnf install pgvector_15
Note: Replace 15
with your Postgres server version
With Conda Postgres, install from conda-forge with:
conda install -c conda-forge pgvector
This method is community-maintained by @mmcauliffe
pgvector is available on these providers.
To request a new extension on other providers:
- Amazon RDS - follow the instructions on this page
- Google Cloud SQL - vote or comment on this page
- Azure Database - vote or comment on this page
- DigitalOcean Managed Databases - vote or comment on this page
- Render - vote or comment on this page
Install the latest version and run:
ALTER EXTENSION vector UPDATE;
If upgrading with Postgres < 13, remove this line from sql/vector--0.3.2--0.4.0.sql
:
ALTER TYPE vector SET (STORAGE = extended);
Then run make install
and ALTER EXTENSION vector UPDATE;
.
If upgrading from 0.2.7 or 0.3.0, recreate all ivfflat
indexes after upgrading to ensure all data is indexed.
-- Postgres 12+
REINDEX INDEX CONCURRENTLY index_name;
-- Postgres < 12
CREATE INDEX CONCURRENTLY temp_name ON table USING ivfflat (column opclass);
DROP INDEX CONCURRENTLY index_name;
ALTER INDEX temp_name RENAME TO index_name;
Thanks to:
- PASE: PostgreSQL Ultra-High-Dimensional Approximate Nearest Neighbor Search Extension
- Faiss: A Library for Efficient Similarity Search and Clustering of Dense Vectors
- Using the Triangle Inequality to Accelerate k-means
- k-means++: The Advantage of Careful Seeding
- Concept Decompositions for Large Sparse Text Data using Clustering
View the changelog
Everyone is encouraged to help improve this project. Here are a few ways you can help:
- Report bugs
- Fix bugs and submit pull requests
- Write, clarify, or fix documentation
- Suggest or add new features
To get started with development:
git clone https://github.com/pgvector/pgvector.git
cd pgvector
make
make install
To run all tests:
make installcheck # regression tests
make prove_installcheck # TAP tests
To run single tests:
make installcheck REGRESS=functions # regression test
make prove_installcheck PROVE_TESTS=test/t/001_wal.pl # TAP test
To enable benchmarking:
make clean && PG_CFLAGS=-DIVFFLAT_BENCH make && make install
Resources for contributors