surface↓classification↓vibes_llm
Classificationdimension · returns varchar

VIBES_LLM

LLM-backed vibe extraction and assignment (escape hatch for VIBES)

Per-row classifier — stable across GROUP BY.

classificationllmllm-escape-hatchscales-largetext

Syntax

VIBES_LLM({{ text }})
VIBES_LLM({{ text }}, {{ num_vibes }})
VIBES_LLM({{ text }}, {{ num_vibes }}, '{{ focus }}')

Arguments

nametypedescription
textVARCHAR—
num_vibesINTEGER—
focusVARCHAR—

About

LLM-backed escape hatch for VIBES. Sends the full collection of texts to the LLM which both invents the vibe categories and assigns each text to its best-matching vibe in a single pass. Cost: O(N) context. Use this only for small collections (<200 rows) or when you specifically need the LLM to make a nuanced assignment judgment that can't be derived from embedding clusters alone. For any non-trivial collection size, prefer VIBES — the hybrid specialist-cluster + O(K) naming version scales arbitrarily. This is a DIMENSION-shaped function for use in GROUP BY clauses.

Examples

LLM escape hatch: vibe extracted

SELECT
  vibes_llm ('This is giving main character energy')

Nearby rabbit holes

same domain
dimension

AUDIENCE

Identify target audience via zero-shot NLI

classificationnlispecialist-zootext+1
dimension

AUDIENCE_LLM

LLM-backed audience identification (escape hatch for AUDIENCE)

classificationllmllm-escape-hatchtext
dimension

AUTHENTICITY

Assess authenticity of content via zero-shot NLI

classificationnlispecialist-zootext
dimension

AUTHENTICITY_LLM

LLM-backed authenticity assessment (escape hatch for AUTHENTICITY)

classificationllmllm-escape-hatchtext
dimension

BUCKET

Classify text into user-specified buckets via zero-shot NLI

classificationnlispecialist-zootext
dimension

BUCKET_LLM

LLM-backed bucketing (escape hatch for BUCKET)

classificationllmllm-escape-hatchtext
Climb back to The Looking Glass