Pure-Rust vector database for the Nubster Data Plane: ANN similarity search, single binary, edge-ready.
EidosDB is a pure-Rust vector index: it receives precomputed embeddings plus payloads and serves approximate nearest neighbor (ANN) similarity search, with no embedded inference and no JVM. It is designed to run as a single binary, deployable at the edge or inside your own infrastructure.
EidosDB is sponsored by Nubster.
Pre-alpha. Heavy development. Not ready for production use.
The public API is unstable and may change between minor versions.
EidosDB does not yet expose a network server. The API is library-level only.
Add the crate to your Cargo.toml:
[dependencies]
eidosdb-core = { git = "https://github.com/nubster-opensources/eidosdb" }Run the benchmarks to verify your setup:
cargo run --release -p eidosdb-benchMost managed vector databases are cloud-only services that require sending your embeddings to a third-party. EidosDB brings ANN search in-process or as a self-hosted binary:
- Pure Rust, no JVM: no garbage collector pauses, predictable latency.
- Single binary: one executable, no sidecar, no daemon manager.
- Edge-ready: low memory footprint, no network required between the index and the caller.
- Hybrid search: combines dense ANN (HNSW) with lexical BM25 via RRF fusion in one query path.
- Schemaless payloads: store arbitrary JSON alongside each vector and filter at query time.
- Not a full-text search engine: BM25 in EidosDB is a complement to dense retrieval, not a replacement for a dedicated search system.
- Not a transactional database: EidosDB does not provide ACID guarantees, foreign keys, or relational joins.
- Not a managed service: there is no hosted tier. You run it yourself.
- Not production-ready yet: the API is pre-1.0 and breaking changes happen between minor versions.
Dedicated guides (architecture overview, HNSW tuning, hybrid search, storage
layer) are not written yet. Until then, the crate-level rustdoc comments are
the source of truth; build them locally with cargo doc --open.
Documentation is a work in progress. Contributions welcome.
See CONTRIBUTING.md.
EidosDB is dual-licensed under the MIT license and the Apache License 2.0, at your option.
Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual-licensed as above, without any additional terms or conditions.
See CONTRIBUTING.md for details, including the Contributor License Agreement (CLA).
Copyright (c) Nubster.