Qdrant · Schema

Mmr

Maximal Marginal Relevance (MMR) algorithm for re-ranking the points.

AIArtificial IntelligenceVector Databases

Properties

Name Type Description
diversity number Tunable parameter for the MMR algorithm. Determines the balance between diversity and relevance. A higher value favors diversity (dissimilarity to selected results), while a lower value favors relevan
candidates_limit integer The maximum number of candidates to consider for re-ranking. If not specified, the `limit` value is used.
View JSON Schema on GitHub

JSON Schema

qdrant-mmr-schema.json Raw ↑
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "#/components/schemas/Mmr",
  "title": "Mmr",
  "description": "Maximal Marginal Relevance (MMR) algorithm for re-ranking the points.",
  "type": "object",
  "properties": {
    "diversity": {
      "description": "Tunable parameter for the MMR algorithm. Determines the balance between diversity and relevance.\n\nA higher value favors diversity (dissimilarity to selected results), while a lower value favors relevance (similarity to the query vector).\n\nMust be in the range [0, 1]. Default value is 0.5.",
      "type": "number",
      "format": "float",
      "maximum": 1,
      "minimum": 0,
      "nullable": true
    },
    "candidates_limit": {
      "description": "The maximum number of candidates to consider for re-ranking.\n\nIf not specified, the `limit` value is used.",
      "type": "integer",
      "format": "uint",
      "maximum": 16384,
      "minimum": 0,
      "nullable": true
    }
  }
}