FastEmbed
A Python library for embedding generation, built on the ONNX Runtime rather than PyTorch and needing no GPU. For the things here it is what turns a chunk into a vector on the machine doing the work rather than through a hosted embedding service.
Where it sits
Used by
Optionally used: the optional local embedding path is installed with the [fastembed] extra; the library itself computes no embeddings the embedding model the index step runs locally, so chunks can be embedded without calling a hosted service
Functional boundary
An early alpha release whose README says to pin the version and test recipes against non-production data first. HANA synchronisation is snapshot-only and does not reconcile deleted rows, so a document for a row deleted in the source stays in the index. A recipe's field allow-list does not decide what the service account may read: the README states that field allow-listing does not replace SAP authorization, and that the package infers no SAP module semantics, authorization scope or safe fields from table names. Only Business Partner has a built-in recipe, every other dataset needs a `KnowledgeRecipe` written for it, and [Qdrant](external:qdrant) local mode is stated to be for development and smaller indexes.
Details
Graph
For your agent
- Ref
external:fastembed- JSON
- /externals/fastembed.json
- MCP
- Connecting an agent