vector-databases
ai-labs/studies/vector-databases/context-v/
8 files here
profiles/7
Chroma — Profile
The hybrid retrieval in Chroma's docs raises `NotImplementedError` on a laptop — `Collection.search()` is gated behind Cloud.
LanceDB — Profile
The database *is* the file format: a `.lance` directory DuckDB, Polars, and Pandas read with no LanceDB process running.
Milvus — Profile
One Milvus deployment is not a process — it is nine roles, three external dependencies, and a C++ ANN core reached through cgo.
pgvector — Profile
About 3k lines of substantive C, and it inherits MVCC, WAL, replication, PITR, and row-level security completely untouched.
Qdrant — Profile
Qdrant estimates a filter's cardinality *before* searching, which is why recall doesn't collapse the moment you filter on `tenant_id`.
Weaviate — Profile
The database itself calls OpenAI to embed on write and Cohere to rerank on read — the RAG pipeline lives inside the store, not the app.
vector-databases/profiles