surface↓summarization↓consensus_llm
Summarizationaggregate · returns varchar

CONSENSUS_LLM

LLM-only consensus (escape hatch for CONSENSUS)

Per-group — reads the whole group in one call.

summarizationllmllm-escape-hatchjson

Syntax

CONSENSUS_LLM({{ texts }})
CONSENSUS_LLM({{ texts }}, '{{ prompt }}')

Arguments

nametypedescription
textsJSON—
prompt(optional)VARCHAR—

About

LLM-only consensus extraction — bypasses the embedding-centrality representative selection used by CONSENSUS and shoves every text into the LLM prompt directly. This is context-window-bound: for small collections (hundreds of texts) it's fine; for larger collections it will either truncate or fail. The canonical CONSENSUS operator handles both shapes transparently — use CONSENSUS_LLM only when you specifically need the LLM to see every text.

Examples

LLM escape hatch finds consensus in positive service reviews

WITH
  test_data AS (
    SELECT
      *
    FROM
      (
        VALUES
          ('The service was excellent'),
          ('Staff was friendly and helpful'),
          ('Great customer service experience'),
          ('Very attentive and professional team')
      ) AS t (review)
  )
SELECT
  CONSENSUS_LLM (review)
FROM
  test_data

Nearby rabbit holes

same domain
scalar

CONDENSE

Condense individual text into brief summary (scalar, per-row)

summarizationllmtext
aggregate

CONSENSUS

Finds common ground among texts via centrality + LLM summary

summarizationllmscales-largejson
aggregate

MERGE_TEXTS

Combine multiple text values into one coherent output

summarizationllmtext
aggregate

SUMMARIZE

Summarize a group of texts into one concise overview

summarizationllmtext
scalar

SUMMARIZE_URLS

Extracts URLs from text, fetches with browser, returns summary

summarizationllmurl
aggregate

THEMES

Extract N main topics from texts (embed centrality + LLM naming)

summarizationllmscales-largejson
Climb back to The Looking Glass