Open-source vector similarity search for Postgres
Store your vectors with the rest of your data. Supports:
- exact and approximate nearest neighbor search
- single-precision, half-precision, binary, and sparse vectors
- L2 distance, inner product, cosine distance, L1 distance, Hamming distance, and Jaccard distance
- any language with a Postgres client
Plus ACID compliance, point-in-time recovery, JOINs, and all of the other great features of Postgres
Compile and install the extension (supports Postgres 12+)
cd /tmp
git clone --branch v0.7.4 https://github.com/pgvector/pgvector.git
cd pgvector
make
make install # may need sudoSee the installation notes if you run into issues
You can also install it with Docker, Homebrew, PGXN, APT, Yum, pkg, or conda-forge, and it comes preinstalled with Postgres.app and many hosted providers. There are also instructions for GitHub Actions.
Ensure C++ support in Visual Studio is installed, and run:
call "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvars64.bat"Note: The exact path will vary depending on your Visual Studio version and edition
Then use nmake to build:
set "PGROOT=C:\Program Files\PostgreSQL\16"
cd %TEMP%
git clone --branch v0.7.4 https://github.com/pgvector/pgvector.git
cd pgvector
nmake /F Makefile.win
nmake /F Makefile.win installSee the installation notes if you run into issues
You can also install it with Docker or conda-forge.
Enable the extension (do this once in each database where you want to use it)
CREATE EXTENSION vector;Create a vector column with 3 dimensions
CREATE TABLE items (id bigserial PRIMARY KEY, embedding vector(3));Insert vectors
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 (<#>), cosine distance (<=>), and L1 distance (<+>, added in 0.7.0)
Note: <#> returns the negative inner product since Postgres only supports ASC order index scans on operators
Create a new table with a vector column
CREATE TABLE items (id bigserial PRIMARY KEY, embedding vector(3));Or add a vector column to an existing table
ALTER TABLE items ADD COLUMN embedding vector(3);Insert vectors
INSERT INTO items (embedding) VALUES ('[1,2,3]'), ('[4,5,6]');Or load vectors in bulk using COPY (example)
COPY items (embedding) FROM STDIN WITH (FORMAT BINARY);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;Get the nearest neighbors to a vector
SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;Supported distance functions are:
<->- L2 distance<#>- (negative) inner product<=>- cosine distance<+>- L1 distance (added in 0.7.0)
Get the nearest neighbors to a row
SELECT * FROM items WHERE id != 1 ORDER BY embedding <-> (SELECT embedding FROM items WHERE id = 1) LIMIT 5;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
Get the distance
SELECT embedding <-> '[3,1,2]' AS distance FROM items;For inner product, multiply by -1 (since <#> returns the negative inner product)
SELECT (embedding <#> '[3,1,2]') * -1 AS inner_product FROM items;For cosine similarity, use 1 - cosine distance
SELECT 1 - (embedding <=> '[3,1,2]') AS cosine_similarity FROM items;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 speed. Unlike typical indexes, you will see different results for queries after adding an approximate index.
Supported index types are:
An HNSW index creates a multilayer graph. It has better query performance than IVFFlat (in terms of speed-recall tradeoff), but has slower build times and uses more memory. Also, an index can be created without any data in the table since there isn’t a training step like IVFFlat.
Add an index for each distance function you want to use.
L2 distance
CREATE INDEX ON items USING hnsw (embedding vector_l2_ops);Note: Use halfvec_l2_ops for halfvec and sparsevec_l2_ops for sparsevec (and similar with the other distance functions)
Inner product
CREATE INDEX ON items USING hnsw (embedding vector_ip_ops);Cosine distance
CREATE INDEX ON items USING hnsw (embedding vector_cosine_ops);L1 distance - added in 0.7.0
CREATE INDEX ON items USING hnsw (embedding vector_l1_ops);Hamming distance - added in 0.7.0
CREATE INDEX ON items USING hnsw (embedding bit_hamming_ops);Jaccard distance - added in 0.7.0
CREATE INDEX ON items USING hnsw (embedding bit_jaccard_ops);Supported types are:
vector- up to 2,000 dimensionshalfvec- up to 4,000 dimensions (added in 0.7.0)bit- up to 64,000 dimensions (added in 0.7.0)sparsevec- up to 1,000 non-zero elements (added in 0.7.0)
Specify HNSW parameters
m- the max number of connections per layer (16 by default)ef_construction- the size of the dynamic candidate list for constructing the graph (64 by default)
CREATE INDEX ON items USING hnsw (embedding vector_l2_ops) WITH (m = 16, ef_construction = 64);A higher value of ef_construction provides better recall at the cost of index build time / insert speed.
Specify the size of the dynamic candidate list for search (40 by default)
SET hnsw.ef_search = 100;A higher value provides better recall at the cost of speed.
Use SET LOCAL inside a transaction to set it for a single query
BEGIN;
SET LOCAL hnsw.ef_search = 100;
SELECT ...
COMMIT;Indexes build significantly faster when the graph fits into maintenance_work_mem
SET maintenance_work_mem = '8GB';A notice is shown when the graph no longer fits
NOTICE: hnsw graph no longer fits into maintenance_work_mem after 100000 tuples
DETAIL: Building will take significantly more time.
HINT: Increase maintenance_work_mem to speed up builds.
Note: Do not set maintenance_work_mem so high that it exhausts the memory on the server
Like other index types, it’s faster to create an index after loading your initial data
Starting with 0.6.0, you can also speed up index creation by increasing the number of parallel workers (2 by default)
SET max_parallel_maintenance_workers = 7; -- plus leaderFor a large number of workers, you may also need to increase max_parallel_workers (8 by default)
Check indexing progress with Postgres 12+
SELECT phase, round(100.0 * blocks_done / nullif(blocks_total, 0), 1) AS "%" FROM pg_stat_progress_create_index;The phases for HNSW are:
initializingloading tuples
An IVFFlat index divides vectors into lists, and then searches a subset of those lists that are closest to the query vector. It has faster build times and uses less memory than HNSW, but has lower query performance (in terms of speed-recall tradeoff).
Three keys to achieving good recall are:
- Create the index after the table has some data
- Choose an appropriate number of lists - a good place to start is
rows / 1000for up to 1M rows andsqrt(rows)for over 1M rows - When querying, specify an appropriate number of probes (higher is better for recall, lower is better for speed) - a good place to start is
sqrt(lists)
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);Note: Use halfvec_l2_ops for halfvec (and similar with the other distance functions)
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);Hamming distance - added in 0.7.0
CREATE INDEX ON items USING ivfflat (embedding bit_hamming_ops) WITH (lists = 100);Supported types are:
vector- up to 2,000 dimensionshalfvec- up to 4,000 dimensions (added in 0.7.0)bit- up to 64,000 dimensions (added in 0.7.0)
Specify the number of probes (1 by default)
SET ivfflat.probes = 10;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 = 10;
SELECT ...
COMMIT;Speed up index creation on large tables by increasing the number of parallel workers (2 by default)
SET max_parallel_maintenance_workers = 7; -- plus leaderFor a large number of workers, you may also need to increase max_parallel_workers (8 by default)
Check indexing progress with Postgres 12+
SELECT phase, round(100.0 * tuples_done / nullif(tuples_total, 0), 1) AS "%" FROM pg_stat_progress_create_index;The phases for IVFFlat are:
initializingperforming k-meansassigning tuplesloading tuples
Note: % is only populated during the loading tuples phase
There are a few ways to index nearest neighbor queries with a WHERE clause
SELECT * FROM items WHERE category_id = 123 ORDER BY embedding <-> '[3,1,2]' LIMIT 5;Create an index on one or more of the WHERE columns for exact search
CREATE INDEX ON items (category_id);Or a partial index on the vector column for approximate search
CREATE INDEX ON items USING hnsw (embedding vector_l2_ops) WHERE (category_id = 123);Use partitioning for approximate search on many different values of the WHERE columns
CREATE TABLE items (embedding vector(3), category_id int) PARTITION BY LIST(category_id);Added in 0.7.0
Use the halfvec type to store half-precision vectors
CREATE TABLE items (id bigserial PRIMARY KEY, embedding halfvec(3));Added in 0.7.0
Index vectors at half precision for smaller indexes
CREATE INDEX ON items USING hnsw ((embedding::halfvec(3)) halfvec_l2_ops);Get the nearest neighbors
SELECT * FROM items ORDER BY embedding::halfvec(3) <-> '[1,2,3]' LIMIT 5;Use the bit type to store binary vectors (example)
CREATE TABLE items (id bigserial PRIMARY KEY, embedding bit(3));
INSERT INTO items (embedding) VALUES ('000'), ('111');Get the nearest neighbors by Hamming distance (added in 0.7.0)
SELECT * FROM items ORDER BY embedding <~> '101' LIMIT 5;Or (before 0.7.0)
SELECT * FROM items ORDER BY bit_count(embedding # '101') LIMIT 5;Also supports Jaccard distance (<%>)
Added in 0.7.0
Use expression indexing for binary quantization
CREATE INDEX ON items USING hnsw ((binary_quantize(embedding)::bit(3)) bit_hamming_ops);Get the nearest neighbors by Hamming distance
SELECT * FROM items ORDER BY binary_quantize(embedding)::bit(3) <~> binary_quantize('[1,-2,3]') LIMIT 5;Re-rank by the original vectors for better recall
SELECT * FROM (
SELECT * FROM items ORDER BY binary_quantize(embedding)::bit(3) <~> binary_quantize('[1,-2,3]') LIMIT 20
) ORDER BY embedding <=> '[1,-2,3]' LIMIT 5;Added in 0.7.0
Use the sparsevec type to store sparse vectors
CREATE TABLE items (id bigserial PRIMARY KEY, embedding sparsevec(5));Insert vectors
INSERT INTO items (embedding) VALUES ('{1:1,3:2,5:3}/5'), ('{1:4,3:5,5:6}/5');The format is {index1:value1,index2:value2}/dimensions and indices start at 1 like SQL arrays
Get the nearest neighbors by L2 distance
SELECT * FROM items ORDER BY embedding <-> '{1:3,3:1,5:2}/5' LIMIT 5;Use together with Postgres full-text search for hybrid search.
SELECT id, content FROM items, plainto_tsquery('hello search') query
WHERE textsearch @@ query ORDER BY ts_rank_cd(textsearch, query) DESC LIMIT 5;You can use Reciprocal Rank Fusion or a cross-encoder to combine results.
Added in 0.7.0
Use expression indexing to index subvectors
CREATE INDEX ON items USING hnsw ((subvector(embedding, 1, 3)::vector(3)) vector_cosine_ops);Get the nearest neighbors by cosine distance
SELECT * FROM items ORDER BY subvector(embedding, 1, 3)::vector(3) <=> subvector('[1,2,3,4,5]'::vector, 1, 3) LIMIT 5;Re-rank by the full vectors for better recall
SELECT * FROM (
SELECT * FROM items ORDER BY subvector(embedding, 1, 3)::vector(3) <=> subvector('[1,2,3,4,5]'::vector, 1, 3) LIMIT 20
) ORDER BY embedding <=> '[1,2,3,4,5]' LIMIT 5;Use a tool like PgTune to set initial values for Postgres server parameters. For instance, shared_buffers should typically be 25% of the server’s memory. You can find the config file with:
SHOW config_file;And check individual settings with:
SHOW shared_buffers;Be sure to restart Postgres for changes to take effect.
Use COPY for bulk loading data (example).
COPY items (embedding) FROM STDIN WITH (FORMAT BINARY);Add any indexes after loading the initial data for best performance.
See index build time for HNSW and IVFFlat.
In production environments, create indexes concurrently to avoid blocking writes.
CREATE INDEX CONCURRENTLY ...Use EXPLAIN ANALYZE to debug performance.
EXPLAIN ANALYZE SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;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 5;To speed up queries with an IVFFlat 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);Vacuuming can take a while for HNSW indexes. Speed it up by reindexing first.
REINDEX INDEX CONCURRENTLY index_name;
VACUUM table_name;Monitor performance with pg_stat_statements (be sure to add it to shared_preload_libraries).
CREATE EXTENSION pg_stat_statements;Get the most time-consuming queries with:
SELECT query, calls, ROUND((total_plan_time + total_exec_time) / calls) AS avg_time_ms,
ROUND((total_plan_time + total_exec_time) / 60000) AS total_time_min
FROM pg_stat_statements ORDER BY total_plan_time + total_exec_time DESC LIMIT 20;Note: Replace total_plan_time + total_exec_time with total_time for Postgres < 13
Monitor recall by comparing results from approximate search with exact search.
BEGIN;
SET LOCAL enable_indexscan = off; -- use exact search
SELECT ...
COMMIT;Scale pgvector the same way you scale Postgres.
Scale vertically by increasing memory, CPU, and storage on a single instance. Use existing tools to tune parameters and monitor performance.
Scale horizontally with replicas, or use Citus or another approach for sharding (example).
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-c |
| C++ | pgvector-cpp |
| C#, F#, Visual Basic | pgvector-dotnet |
| Crystal | pgvector-crystal |
| Dart | pgvector-dart |
| Elixir | pgvector-elixir |
| Go | pgvector-go |
| Haskell | pgvector-haskell |
| Java, Kotlin, Groovy, Scala | pgvector-java |
| JavaScript, TypeScript | pgvector-node |
| Julia | pgvector-julia |
| Lisp | pgvector-lisp |
| Lua | pgvector-lua |
| Nim | pgvector-nim |
| OCaml | pgvector-ocaml |
| Perl | pgvector-perl |
| PHP | pgvector-php |
| Python | pgvector-python |
| R | pgvector-r |
| Ruby | pgvector-ruby, Neighbor |
| Rust | pgvector-rust |
| Swift | pgvector-swift |
| Zig | pgvector-zig |
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.
You can use half-precision indexing to index up to 4,000 dimensions or binary quantization to index up to 64,000 dimensions. Another option is dimensionality reduction.
You can use vector as the type (instead of vector(3)).
CREATE TABLE embeddings (model_id bigint, item_id bigint, embedding vector, PRIMARY KEY (model_id, item_id));However, you can only create indexes on rows with the same number of dimensions (using expression and partial indexing):
CREATE INDEX ON embeddings USING hnsw ((embedding::vector(3)) vector_l2_ops) WHERE (model_id = 123);and query with:
SELECT * FROM embeddings WHERE model_id = 123 ORDER BY embedding::vector(3) <-> '[3,1,2]' LIMIT 5;You can use the double precision[] or numeric[] type to store vectors with more precision.
CREATE TABLE items (id bigserial PRIMARY KEY, embedding double precision[]);
-- use {} instead of [] for Postgres arrays
INSERT INTO items (embedding) VALUES ('{1,2,3}'), ('{4,5,6}');Optionally, add a check constraint to ensure data can be converted to the vector type and has the expected dimensions.
ALTER TABLE items ADD CHECK (vector_dims(embedding::vector) = 3);Use expression indexing to index (at a lower precision):
CREATE INDEX ON items USING hnsw ((embedding::vector(3)) vector_l2_ops);and query with:
SELECT * FROM items ORDER BY embedding::vector(3) <-> '[3,1,2]' LIMIT 5;No, but like other index types, you’ll likely see better performance if they do. You can get the size of an index with:
SELECT pg_size_pretty(pg_relation_size('index_name'));The query needs to have an ORDER BY and LIMIT, and the ORDER BY must be the result of a distance operator (not an expression) in ascending order.
-- index
ORDER BY embedding <=> '[3,1,2]' LIMIT 5;
-- no index
ORDER BY 1 - (embedding <=> '[3,1,2]') DESC LIMIT 5;You can encourage the planner to use an index for a query with:
BEGIN;
SET LOCAL enable_seqscan = off;
SELECT ...
COMMIT;Also, if the table is small, a table scan may be faster.
The planner doesn’t consider out-of-line storage in cost estimates, which can make a serial scan look cheaper. You can reduce the cost of a parallel scan for a query with:
BEGIN;
SET LOCAL min_parallel_table_scan_size = 1;
SET LOCAL parallel_setup_cost = 1;
SELECT ...
COMMIT;or choose to store vectors inline:
ALTER TABLE items ALTER COLUMN embedding SET STORAGE PLAIN;Results are limited by the size of the dynamic candidate list (hnsw.ef_search). There may be even less results due to dead tuples or filtering conditions in the query. We recommend setting hnsw.ef_search to at least twice the LIMIT of the query. If you need more than 500 results, use an IVFFlat index instead.
Also, note that NULL vectors are not indexed (as well as zero vectors for cosine distance).
The index was likely created with too little data for the number of lists. Drop the index until the table has more data.
DROP INDEX index_name;Results can also be limited by the number of probes (ivfflat.probes).
Also, note that NULL vectors are not indexed (as well as zero vectors for cosine distance).
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 | Added |
|---|---|---|
| + | element-wise addition | |
| - | element-wise subtraction | |
| * | element-wise multiplication | 0.5.0 |
| || | concatenate | 0.7.0 |
| <-> | Euclidean distance | |
| <#> | negative inner product | |
| <=> | cosine distance | |
| <+> | taxicab distance | 0.7.0 |
| Function | Description | Added |
|---|---|---|
| binary_quantize(vector) → bit | binary quantize | 0.7.0 |
| cosine_distance(vector, vector) → double precision | cosine distance | |
| inner_product(vector, vector) → double precision | inner product | |
| l1_distance(vector, vector) → double precision | taxicab distance | 0.5.0 |
| l2_distance(vector, vector) → double precision | Euclidean distance | |
| l2_normalize(vector) → vector | Normalize with Euclidean norm | 0.7.0 |
| subvector(vector, integer, integer) → vector | subvector | 0.7.0 |
| vector_dims(vector) → integer | number of dimensions | |
| vector_norm(vector) → double precision | Euclidean norm |
| Function | Description | Added |
|---|---|---|
| avg(vector) → vector | average | |
| sum(vector) → vector | sum | 0.5.0 |
Each half vector takes 2 * dimensions + 8 bytes of storage. Each element is a half-precision floating-point number, and all elements must be finite (no NaN, Infinity or -Infinity). Half vectors can have up to 16,000 dimensions.
| Operator | Description | Added |
|---|---|---|
| + | element-wise addition | 0.7.0 |
| - | element-wise subtraction | 0.7.0 |
| * | element-wise multiplication | 0.7.0 |
| || | concatenate | 0.7.0 |
| <-> | Euclidean distance | 0.7.0 |
| <#> | negative inner product | 0.7.0 |
| <=> | cosine distance | 0.7.0 |
| <+> | taxicab distance | 0.7.0 |
| Function | Description | Added |
|---|---|---|
| binary_quantize(halfvec) → bit | binary quantize | 0.7.0 |
| cosine_distance(halfvec, halfvec) → double precision | cosine distance | 0.7.0 |
| inner_product(halfvec, halfvec) → double precision | inner product | 0.7.0 |
| l1_distance(halfvec, halfvec) → double precision | taxicab distance | 0.7.0 |
| l2_distance(halfvec, halfvec) → double precision | Euclidean distance | 0.7.0 |
| l2_norm(halfvec) → double precision | Euclidean norm | 0.7.0 |
| l2_normalize(halfvec) → halfvec | Normalize with Euclidean norm | 0.7.0 |
| subvector(halfvec, integer, integer) → halfvec | subvector | 0.7.0 |
| vector_dims(halfvec) → integer | number of dimensions | 0.7.0 |
| Function | Description | Added |
|---|---|---|
| avg(halfvec) → halfvec | average | 0.7.0 |
| sum(halfvec) → halfvec | sum | 0.7.0 |
Each bit vector takes dimensions / 8 + 8 bytes of storage. See the Postgres docs for more info.
| Operator | Description | Added |
|---|---|---|
| <~> | Hamming distance | 0.7.0 |
| <%> | Jaccard distance | 0.7.0 |
| Function | Description | Added |
|---|---|---|
| hamming_distance(bit, bit) → double precision | Hamming distance | 0.7.0 |
| jaccard_distance(bit, bit) → double precision | Jaccard distance | 0.7.0 |
Each sparse vector takes 8 * non-zero elements + 16 bytes of storage. Each element is a single-precision floating-point number, and all elements must be finite (no NaN, Infinity or -Infinity). Sparse vectors can have up to 16,000 non-zero elements.
| Operator | Description | Added |
|---|---|---|
| <-> | Euclidean distance | 0.7.0 |
| <#> | negative inner product | 0.7.0 |
| <=> | cosine distance | 0.7.0 |
| <+> | taxicab distance | 0.7.0 |
| Function | Description | Added |
|---|---|---|
| cosine_distance(sparsevec, sparsevec) → double precision | cosine distance | 0.7.0 |
| inner_product(sparsevec, sparsevec) → double precision | inner product | 0.7.0 |
| l1_distance(sparsevec, sparsevec) → double precision | taxicab distance | 0.7.0 |
| l2_distance(sparsevec, sparsevec) → double precision | Euclidean distance | 0.7.0 |
| l2_norm(sparsevec) → double precision | Euclidean norm | 0.7.0 |
| l2_normalize(sparsevec) → sparsevec | Normalize with Euclidean norm | 0.7.0 |
If your machine has multiple Postgres installations, specify the path to pg_config with:
export PG_CONFIG=/Library/PostgreSQL/16/bin/pg_configThen re-run the installation instructions (run make clean before make if needed). If sudo is needed for make install, use:
sudo --preserve-env=PG_CONFIG make installA few common paths on Mac are:
- EDB installer -
/Library/PostgreSQL/16/bin/pg_config - Homebrew (arm64) -
/opt/homebrew/opt/postgresql@16/bin/pg_config - Homebrew (x86-64) -
/usr/local/opt/postgresql@16/bin/pg_config
Note: Replace 16 with your Postgres server version
If compilation fails with fatal error: postgres.h: No such file or directory, make sure Postgres development files are installed on the server.
For Ubuntu and Debian, use:
sudo apt install postgresql-server-dev-16Note: Replace 16 with your Postgres server version
If compilation fails and the output includes warning: no such sysroot directory on Mac, reinstall Xcode Command Line Tools.
By default, pgvector compiles with -march=native on some platforms for best performance. However, this can lead to Illegal instruction errors if trying to run the compiled extension on a different machine.
To compile for portability, use:
make OPTFLAGS=""If compilation fails with Cannot open include file: 'postgres.h': No such file or directory, make sure PGROOT is correct.
If installation fails with Access is denied, re-run the installation instructions as an administrator.
Get the Docker image with:
docker pull pgvector/pgvector:pg16This adds pgvector to the Postgres image (replace 16 with your Postgres server version, and run it the same way).
You can also build the image manually:
git clone --branch v0.7.4 https://github.com/pgvector/pgvector.git
cd pgvector
docker build --pull --build-arg PG_MAJOR=16 -t myuser/pgvector .With Homebrew Postgres, you can use:
brew install pgvectorNote: This only adds it to the postgresql@14 formula
Install from the PostgreSQL Extension Network with:
pgxn install vectorDebian and Ubuntu packages are available from the PostgreSQL APT Repository. Follow the setup instructions and run:
sudo apt install postgresql-16-pgvectorNote: Replace 16 with your Postgres server version
RPM packages are available from the PostgreSQL Yum Repository. Follow the setup instructions for your distribution and run:
sudo yum install pgvector_16
# or
sudo dnf install pgvector_16Note: Replace 16 with your Postgres server version
Install the FreeBSD package with:
pkg install postgresql15-pgvectoror the port with:
cd /usr/ports/databases/pgvector
make installWith Conda Postgres, install from conda-forge with:
conda install -c conda-forge pgvectorThis method is community-maintained by @mmcauliffe
Download the latest release with Postgres 15+.
pgvector is available on these providers.
Install the latest version (use the same method as the original installation). Then in each database you want to upgrade, run:
ALTER EXTENSION vector UPDATE;You can check the version in the current database with:
SELECT extversion FROM pg_extension WHERE extname = 'vector';If upgrading with Postgres 12, remove this line from sql/vector--0.5.1--0.6.0.sql:
ALTER TYPE vector SET (STORAGE = external);Then run make install and ALTER EXTENSION vector UPDATE;.
The Docker image is now published in the pgvector org, and there are tags for each supported version of Postgres (rather than a latest tag).
docker pull pgvector/pgvector:pg16
# or
docker pull pgvector/pgvector:0.6.0-pg16Also, if you’ve increased maintenance_work_mem, make sure --shm-size is at least that size to avoid an error with parallel HNSW index builds.
docker run --shm-size=1g ...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
- Efficient and Robust Approximate Nearest Neighbor Search using Hierarchical Navigable Small World Graphs
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 installTo run all tests:
make installcheck # regression tests
make prove_installcheck # TAP testsTo run single tests:
make installcheck REGRESS=functions # regression test
make prove_installcheck PROVE_TESTS=test/t/001_ivfflat_wal.pl # TAP testTo enable assertions:
make clean && PG_CFLAGS="-DUSE_ASSERT_CHECKING" make && make installTo enable benchmarking:
make clean && PG_CFLAGS="-DIVFFLAT_BENCH" make && make installTo show memory usage:
make clean && PG_CFLAGS="-DHNSW_MEMORY -DIVFFLAT_MEMORY" make && make installTo get k-means metrics:
make clean && PG_CFLAGS="-DIVFFLAT_KMEANS_DEBUG" make && make installResources for contributors
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent