pg_turbovec
Overview
| Package | Version | Category | License | Language |
|---|---|---|---|---|
pg_turbovec | 1.29.0 | RAG | Apache-2.0 | Rust |
| ID | Extension | Bin | Lib | Load | Create | Trust | Reloc | Schema |
|---|---|---|---|---|---|---|---|---|
| 1980 | pg_turbovec | No | Yes | No | Yes | No | No | turbovec |
| Related | pg_turboquant vector vchord vectorscale |
|---|
Upstream v1.29.0 supports PostgreSQL 13-18 with PG19 beta experimental; PIGSTY RPM, DEB, and source remain at 1.28.3 for PostgreSQL 14-18 with pgrx 0.19.1 and OpenBLAS. Upstream 1.28.4 fixes a persisted row-count corruption bug present in 1.28.3.
Version
| Type | Repo | Version | PG Ver | Package | Deps |
|---|---|---|---|---|---|
| EXT | PIGSTY | 1.29.0 | 1817161514 | pg_turbovec | - |
| RPM | PIGSTY | 1.28.3 | 1817161514 | pg_turbovec_$v | openblas |
| DEB | PIGSTY | 1.28.3 | 1817161514 | postgresql-$v-pg-turbovec | libopenblas0 |
Build
You can build the RPM / DEB packages for pg_turbovec using pig build:
pig build pkg pg_turbovec # build RPM / DEB packages
Install
You can install pg_turbovec directly. First, make sure the PGDG and PIGSTY repositories are added and enabled:
pig repo add pgsql -u # Add repo and update cache
Install the extension using pig or apt/yum/dnf:
pig install pg_turbovec; # Install for current active PG version
pig ext install -y pg_turbovec -v 18 # PG 18
pig ext install -y pg_turbovec -v 17 # PG 17
pig ext install -y pg_turbovec -v 16 # PG 16
pig ext install -y pg_turbovec -v 15 # PG 15
pig ext install -y pg_turbovec -v 14 # PG 14
dnf install -y pg_turbovec_18 # PG 18
dnf install -y pg_turbovec_17 # PG 17
dnf install -y pg_turbovec_16 # PG 16
dnf install -y pg_turbovec_15 # PG 15
dnf install -y pg_turbovec_14 # PG 14
apt install -y postgresql-18-pg-turbovec # PG 18
apt install -y postgresql-17-pg-turbovec # PG 17
apt install -y postgresql-16-pg-turbovec # PG 16
apt install -y postgresql-15-pg-turbovec # PG 15
apt install -y postgresql-14-pg-turbovec # PG 14
Create Extension:
CREATE EXTENSION pg_turbovec;
Usage
Sources:
- pg_turbovec v1.29.0 README
- pg_turbovec v1.29.0 changelog
- pg_turbovec v1.29.0 control file
- Partitioned-scale guide
- Pigsty package matrix
pg_turbovec provides a dense turbovec.vector type and a turbovec approximate-nearest-neighbor index access method. It quantizes floating-point coordinates to 2, 3, or 4 bits and reranks candidates against heap vectors. Use it when index footprint and search throughput matter more than exact nearest-neighbor recall.
Create and Query Vectors
CREATE EXTENSION pg_turbovec;
SET search_path = public, turbovec;
CREATE TABLE items (
id bigint GENERATED ALWAYS AS IDENTITY PRIMARY KEY,
embedding turbovec.vector
);
INSERT INTO items (embedding)
VALUES ('[1,2,3]'), ('[4,5,6]');
SELECT id, embedding <=> '[3,1,2]'::turbovec.vector AS cosine_distance
FROM items
ORDER BY embedding <=> '[3,1,2]'::turbovec.vector
LIMIT 10;
Distance operators are <-> for L2, <#> for negative inner product, <=> for cosine distance, and <+> for L1. The current index supports inner-product and cosine ordering; L2 and L1 are exact-only operations.
Build a TurboVec Index
CREATE INDEX items_embedding_idx ON items
USING turbovec (embedding vec_cosine_ops)
WITH (bit_width = 4);
SET turbovec.probes = 32;
SELECT id
FROM items
ORDER BY embedding <=> '[3,1,2]'::turbovec.vector
LIMIT 10;
Use vec_cosine_ops with <=> or vec_ip_ops with <#>. bit_width = 4 is the default and generally favors recall; 2-bit indexes are smaller but need workload-specific recall testing. CREATE INDEX CONCURRENTLY is supported.
Important tuning controls include turbovec.probes, turbovec.search_k, turbovec.oversample, turbovec.hi_dim_rerank, turbovec.iterative_scan, and turbovec.cache_size_mb. Change one dimension at a time and compare approximate results with an exact baseline.
Filtering and Partitioning
Use PostgreSQL partial indexes for stable filter values, the documented turbovec.knn(..., allowed) surface for an explicit candidate allowlist, or iterative scan for normal filtered ORDER BY ... LIMIT queries.
Version 1.29 documents native PostgreSQL partitioning for larger-than-single-table datasets. A parent query can use Merge Append across per-partition TurboVec indexes:
SELECT id
FROM partitioned_items
ORDER BY embedding <=> $1::turbovec.vector
LIMIT 20;
Build, vacuum, and reindex each partition independently. Partition pruning based on a coarse vector quantizer is only a design in 1.29.0, not a shipped feature.
Version and Integrity Boundaries
- The control file installs objects in schema
turbovec, is not relocatable, and does not requireshared_preload_librariesor a server restart. - Upstream v1.29 targets PostgreSQL 13-18 and labels PostgreSQL 19 support experimental; current Pigsty packages cover PostgreSQL 14-18 and provide the matching OpenBLAS-linked binary.
- The current Pigsty package/source metadata remains at
1.28.3, while upstream is1.29.0. On a packaged 1.28.3 installation, do not assume the 1.28.4 integrity checker or 1.29 features are present. - Upstream 1.28.4 fixes persisted row-count drift that could corrupt the index ID table and adds
turbovec.turbovec_check(regclass). Prioritize a package upgrade to at least 1.28.4 before write-heavy production use. An already corrupt index still needsREINDEXor drop/recreate recovery. - Version 1.29.0 is additive and does not require reindexing when upgrading from a healthy 1.28.4 index.
ALTER EXTENSION pg_turbovec UPDATE TO '1.29.0'is sufficient after the new files are installed. - Although the 1.29 reloption parser accepts
bit_width = 1, end-to-end one-bit indexing is not implemented andCREATE INDEXintentionally errors. Use 2, 3, or 4. - The on-disk ID table still has a documented crash-safety gap after an unclean shutdown. Treat integrity errors as actionable and follow the upstream recovery guidance.
Feedback
Was this page helpful?
Thanks for the feedback! Please let us know how we can improve.
Sorry to hear that. Please let us know how we can improve.