$ memista

memista vs USearch (direct)

updated 2026-06-01 · grounded in source

memista uses USearch as its vector index. The honest comparison is therefore not “library A vs library B” — it is “do I want USearch by itself, or with the metadata layer, persistence helpers, and HTTP service memista wraps around it?“

dimensionmemista 0.1.xUSearch 2.19.x (direct)
LayerLibrary + bundled HTTP serverLibrary only
Vector indexUSearch (transitive)USearch (native)
Metric availableIP (hardcoded in IndexOptions)IP, Cosine, L2, Hamming, Tanimoto, Sorensen, Haversine, Divergence
QuantizationF32 (hardcoded)F32, F16, F64, BF16, I8, B1
Embedding dimensionshardcoded to 2 in current crateruntime-set
Metadata storageSQLite (chunks_<db_id> table)none — you bring your own
Multi-tenantby database_id (one table + one index file each)by separate index instances
HTTP API/v1/insert, /v1/search, /v1/drop via Actix-webnone
OpenAPI / docs UIyes — apistos + Swagger / Redoc / RapiDoc / Scalarnone
PersistenceSQLite WAL + <db_id>.usearch file.usearch file (you call save / load)
Asyncyes (tokio, async-sqlite, actix-web)sync core, async wrappers available
Tuning surfaceaccepts USearch defaults (connectivity: 0, expansion_*: 0)full IndexOptions exposed
Statusexperimental (per README)production at Unum
LicenseGPL-3.0Apache-2.0

When to pick USearch directly

When to pick memista

What memista does not do that USearch does

What USearch does not do that memista does

The honest summary

memista is a convenience wrapper. If you want the convenience — SQLite plus HTTP plus OpenAPI plus partitioning, in one crate — it earns its keep. If you want the full power of USearch and don’t need a service shape, use USearch directly. The two are not really competitors; memista is a configuration of USearch with batteries.

Sources: src/lib.rs and Cargo.toml in this repository; USearch documentation at https://github.com/unum-cloud/usearch.

← all comparisons