surface↓embedding↓embed
Embeddingscalar · returns double[]

EMBED

Generate 768-dim embedding vector from text (on-box nomic-embed-text-v1.5)

Per-row — runs once for each row.

embeddingembedding-modelspecialist-zooscales-largenumeric

Arguments

nametypedescription
textVARCHARText to embed
model(optional)VARCHAROptional embedding model name

About

Generate text embeddings using Agent.embed(). Routes through the specialist zoo gateway (bge-m3 via TEI, 1024 dims, 8192 token context) — see lars.model_defaults.DEFAULT_EMBEDDING_MODEL. Results are cached by input hash for performance. SQL Usage: SELECT id, text, semantic_embed(text) as embedding FROM documents; Performance: - GPU-batched through the zoo: 2000+ texts/sec under batch load - Single-call latency: 5-20ms per text over LAN to the zoo gateway - Cached calls: <1ms (instant return)

Nearby rabbit holes

same domain
scalar

EMBED_BATCH

Batch embed rows and store in lars_embeddings

embeddingembedding-modelspecialist-zootext
scalar

EMBED_BATCH_ELASTIC

Batch embed rows and store in Elasticsearch for hybrid search

embeddingembedding-modelspecialist-zootext
scalar

EMBED_BATCH_PINECONE

Batch embed rows and store in Pinecone for vector search

embeddingembedding-modelspecialist-zooscales-large+1
scalar

EMBED_STATUS

Check embedding coverage for a table/column

embeddingembedding-modelspecialist-zootext
scalar

EMBED_WITH_STORAGE

Generate embedding and store with table/column/ID tracking

embeddingembedding-modelspecialist-zootext
scalar

IMAGE_EMBED

SigLIP 2 embedding for an image (L2-normalized, shared image/text space)

embeddingembedding-modelspecialist-zooimage
Climb back to The Looking Glass