AI Gateway · Component · vector-search-embedding
Vector storage, embedding and rerank
What it does
Manage vector indexes/collections, upsert/query vectors with filters and invoke supported embedding/rerank operations.
Core pattern
Filtered vector storage + similarity candidate queries
- Input
- Documents/vectors or query
- Output
- Stored vectors, nearest neighbors or reranked candidates
Key parameters
| Name | Type | Default | Values |
|---|---|---|---|
| index_collection | — | — | — |
| query | string | — | — |
| filter | — | — | — |
| top_k | integer | 20 | 1–1000 |
| embedding_model | — | — | — |
Maturity and limits
Vector lifecycle tests and explicit capability-matrix coverage. Static source inspection only; no runtime, provider, deployment or test-pass claim.
config with credentials present. Apache-2.0; provider terms apply. BYOK and operation scopes required. Live-ready labels are repository claims, not revalidated here; deferred/catalog-only operations are not normal callable features.
Used in pipelines
Industry tags
- Legal servicesTransferable use
For clause lookup in an authorized contract library, use filtered vector storage followed by similarity candidate queries.
Skills pack
Contract scaffold and deterministic mock
Download · 11KBSHA-256 6fe420f491244632dc3715caf3a02e4d0f483ca4249c856429c87d1bfcea751d
Related components
-
Image generation, editing, upscaling and prediction jobs
Image generation or masked edits + async result tracking
Working 1 pipeline -
Speech transcription and audio translation adapters
Audio transcription or translation + language and format controls
Working 1 pipeline -
Asynchronous video generation and retained media
Async media requests + result polling and usage records
Working 1 pipeline -
Web search and research synthesis
Search request aggregation + source-linked research output
Working 1 pipeline