surface↓similarity↓sounds_like_llm
Similarityscalar · returns boolean

SOUNDS_LIKE_LLM

LLM-backed phonetic similarity (escape hatch for SOUNDS_LIKE)

Per-row — runs once for each row.

similarityllmllm-escape-hatchtext

Syntax

{{ text }} SOUNDS_LIKE_LLM {{ reference }}

Arguments

nametypedescription
textVARCHARFirst text
referenceVARCHARReference text

About

LLM-backed escape hatch for SOUNDS_LIKE. Use when the deterministic jellyfish-based matcher in SOUNDS_LIKE can't make a confident call — typically: - Cross-script transliterations (e.g., "Mohammed" vs "محمد") - Cross-language phonetics (e.g., Japanese romaji variants) - Very rare or archaic name spellings - Cases where context matters (nicknames, diminutives) For standard English-name phonetic matching (Smith/Smyth, Catherine/Kathryn, Stephen/Steven, etc.), prefer SOUNDS_LIKE — it's pure Python with no GPU/LLM cost and scores 23/23 on the calibration set. SQL Usage: SELECT * FROM customers WHERE name SOUNDS_LIKE_LLM 'Mohammed';

Examples

LLM escape hatch: phonetically similar names

SELECT
  sounds_like_llm ('Smith', 'Smyth')

Nearby rabbit holes

same domain
scalar

ALIGNS

How strongly text supports a specific message or stance (0.0-1.0)

similarityllmtext
scalar

IMAGE_MATCHES

Check if an image semantically matches a text query (cross-modal)

similarityembedding-modelspecialist-zooimage
scalar

IMAGE_SIMILARITY

Cross-modal cosine similarity between an image and a text query

similarityembedding-modelspecialist-zooimage
scalar

MATCH

Fuzzy cross-match two string arrays via bge-m3 embeddings

similarityllmjson
scalar

MATCH_PAIR

Checks whether two values match under a relationship

similarityembedding-modelspecialist-zootext
scalar

MEANS

Returns TRUE if text semantically matches the criterion (cross-encoder)

similarityrerankerspecialist-zootext
Climb back to The Looking Glass