Vector icons

The 20 vector icons in RAG & Search, described

ANN searchann-search

ANN search — approximate nearest neighbours, close enough and found fast.

Cosine similaritycosine

Cosine similarity — the angle between two vectors as a closeness score.

Dimension reducedimension-reduce

Dimensionality reduction — the same shape in fewer numbers, projected smaller.

Embed batchembed-batch

Embed batch — many texts through the embedding model at once for throughput.

Embed cacheembed-cache

An embed cache — vectors computed once and kept for the next time the same text appears.

Embeddingembedding

A vector embedding — a piece of meaning turned into coordinates a model can compare and search.

Exact searchexact-search

Exact search — not close, the one precise match for the exact term.

HNSWhnsw

HNSW — a layered graph index you can search for nearest neighbours quickly.

Flat indexindex-flat

A flat index — compare the query against every single vector, exact but slow.

IVF indexindex-ivf

An IVF index — ask the nearest cluster rather than the whole world, approximate and fast.

Index mergeindex-merge

Index merge — two segments compacted and combined into one bigger index.

Index rebuildindex-rebuild

Index rebuild — the index made again from scratch in a batch reindex.

Warm indexindex-warm

A warm index — kept loaded and hot so the first query is fast.

Matryoshkamatryoshka

Matryoshka embeddings — the small vector nests inside the big one, so you can truncate dimensions.

Re-embedreembed

Re-embed — a new model over the same corpus, every vector computed again.

Sharded indexshard-index

A sharded index — the index cut into pieces so many machines can hold it.

Vector databasevector-database

A vector database — stores embeddings and finds the nearest neighbours by similarity.

Vector indexvector-index

A vector index — embeddings filed for finding, an approximate nearest-neighbour store.

Vector queryvector-query

A vector query — asking the index what is nearest to this embedding.

Vector upsertvector-upsert

Vector upsert — write new or updated vectors into the index in one operation.

In code, each is one import — import { AnnSearch } from "@iconmind/react/icons/ann-search" — and the same name in Vue, Svelte, Solid, Preact, React Native, Astro, Blade and Flutter.

Narrow it by tag