post-cortex-embeddings 0.3.0

Embedding engines (Model2Vec static + BERT via Candle + static-hash fallback) and HNSW vector database for post-cortex. Self-contained ML stack — depend on this crate when you need vector search without the full post-cortex orchestrator.

post-cortex-embeddings

Embedding engines + HNSW vector database for post-cortex.

Self-contained ML stack — anyone needing a Candle-backed BERT embedder or an HNSW index for nearest-neighbour search can depend on this crate without pulling the post-cortex daemon or orchestrator. Implements [EmbeddingBackend] (BERT via Candle, static-hash fallback) and ships [VectorDB] (HNSW with optional product quantisation).

Install

[dependencies]
post-cortex-embeddings = "0.2"           # full BERT + HNSW (default)
# or, if you only need the data types (VectorMetadata, SearchMatch, ...):
post-cortex-embeddings = { version = "0.2", default-features = false }

Features

Feature	Default	What it enables
`bert`	yes	Candle + Tokenizers + hf-hub for the BERT backend
`otel`	no	OpenTelemetry instrumentation hooks

When bert is off the crate compiles only vector_db::types — useful for storage backends that need VectorMetadata round-tripping without pulling the ML stack.

Example

use post_cortex_embeddings::{
    LocalEmbeddingEngine, EmbeddingConfig, VectorDB, VectorDbConfig, VectorMetadata,
};
use std::sync::Arc;

# async fn run() -> Result<(), Box<dyn std::error::Error>> {
let engine = Arc::new(LocalEmbeddingEngine::new(EmbeddingConfig::default()).await?);
let db = Arc::new(VectorDB::new(VectorDbConfig::default()));

let vector = engine.embed("hello world").await?;
let meta = VectorMetadata::new(
    "doc-1".into(),
    "hello world".into(),
    "session-1".into(),
    "UpdateContent".into(),
);
db.add_vector(1, vector.clone(), meta).await?;

let hits = db.search(&vector, 5).await?;
for hit in hits {
    println!("{} (similarity {:.3})", hit.metadata.text, hit.similarity);
}
# Ok(()) }

License

MIT — see LICENSE.