# agentsea_embeddings v0.1.0 - Table of Contents

> AgentSea embeddings: embedder/vector-store behaviours, RAG, and in-memory/pgvector/Qdrant/Pinecone stores.

## Modules

- [AgentSea.Embedder](AgentSea.Embedder.md): Turns text into vectors. Adapters: the dependency-free
`AgentSea.Embedder.Hashing` (good for tests/dev) and, in future, Bumblebee/Nx
(in-process HF/ONNX models) or remote embedding providers.

- [AgentSea.Embedder.Cohere](AgentSea.Embedder.Cohere.md): Cohere embeddings adapter (`POST /v1/embed`) over `Req` — a remote
`AgentSea.Embedder`.
- [AgentSea.Embedder.Hashing](AgentSea.Embedder.Hashing.md): A deterministic, dependency-free embedder using the hashing trick: tokens are
hashed into fixed-dimension buckets (bag-of-words), then the vector is L2
normalized. Texts that share words land closer together — enough for tests,
local dev, and demos without pulling in an ML runtime.

- [AgentSea.Embedder.OpenAI](AgentSea.Embedder.OpenAI.md): OpenAI embeddings adapter (`POST /v1/embeddings`) over `Req` — a remote
`AgentSea.Embedder` (no local model).
- [AgentSea.Embeddings](AgentSea.Embeddings.md): Ties an `AgentSea.Embedder` to an `AgentSea.VectorStore`: embed-and-index text
documents, then semantic-search by text.
- [AgentSea.Embeddings.RetrievalTool](AgentSea.Embeddings.RetrievalTool.md): An `AgentSea.Tool` that lets an agent search a knowledge base — the retrieval
half of RAG.
- [AgentSea.Memory.Vector](AgentSea.Memory.Vector.md): Vector-backed conversation memory: messages are embedded and indexed, so
`search/2` recalls the most *relevant* past messages (not just the most
recent). Wraps an `AgentSea.Embeddings` handle (any embedder + vector store).
- [AgentSea.Vector](AgentSea.Vector.md): Small vector math: dot product, L2 norm, normalization, cosine similarity.
- [AgentSea.VectorStore](AgentSea.VectorStore.md): Stores vectors and answers nearest-neighbour queries. Adapters: the in-memory
`AgentSea.VectorStore.Memory` and, in future, pgvector (first-class via Ecto)
or remote stores (Pinecone/Qdrant) over HTTP.

- [AgentSea.VectorStore.Memory](AgentSea.VectorStore.Memory.md): In-memory vector store backed by a `GenServer`. Brute-force cosine-similarity
search — fine for tests, small corpora, and demos. Records are keyed by id.

- [AgentSea.VectorStore.Pinecone](AgentSea.VectorStore.Pinecone.md): A [Pinecone](https://pinecone.io) `AgentSea.VectorStore` over its data-plane
REST API (`Req`) — a managed/remote store alongside the in-memory, pgvector,
and Qdrant stores.
- [AgentSea.VectorStore.Postgres](AgentSea.VectorStore.Postgres.md): A pgvector-backed `AgentSea.VectorStore` over Postgrex — the design's first-
class production store.
- [AgentSea.VectorStore.Qdrant](AgentSea.VectorStore.Qdrant.md): A [Qdrant](https://qdrant.tech) `AgentSea.VectorStore` over its REST API (`Req`)
— a managed/remote alternative to the in-memory and pgvector stores.

