Ranking score

Learn how Meilisearch computes the _rankingScore for each document and which index settings influence it.

The _rankingScore is a normalized value between 0.0 and 1.0 that represents how relevant a document is to a given search query. A score of 1.0 means the document is a perfect match, while a score closer to 0.0 means it is a weak match. Meilisearch does not return the ranking score by default; you must explicitly request it.

Requesting the ranking score

To include _rankingScore in search results, set showRankingScore to true in your search request:

curl \
  -X POST 'MEILISEARCH_URL/indexes/movies/search' \
  -H 'Content-Type: application/json' \
  --data-binary '{
    "q": "batman dark knight",
    "showRankingScore": true
  }'

Each document in the response will include a _rankingScore field:

{
  "hits": [
    {
      "id": 155,
      "title": "The Dark Knight",
      "_rankingScore": 0.9546
    },
    {
      "id": 36657,
      "title": "Batman Begins",
      "_rankingScore": 0.8103
    }
  ],
  "query": "batman dark knight"
}

Requesting a detailed breakdown

For a deeper understanding of why a document received a particular score, set showRankingScoreDetails to true. This returns the contribution of each ranking rule:

curl \
  -X POST 'MEILISEARCH_URL/indexes/movies/search' \
  -H 'Content-Type: application/json' \
  --data-binary '{
    "q": "batman dark knight",
    "showRankingScore": true,
    "showRankingScoreDetails": true
  }'

The response includes a _rankingScoreDetails object for each document:

{
  "hits": [
    {
      "id": 155,
      "title": "The Dark Knight",
      "_rankingScore": 0.9546,
      "_rankingScoreDetails": {
        "words": { "order": 0, "matchingWords": 3, "maxMatchingWords": 3, "score": 1.0 },
        "typo": { "order": 1, "typoCount": 0, "maxTypoCount": 3, "score": 1.0 },
        "proximity": { "order": 2, "score": 0.9286 },
        "attributeRank": { "order": 3, "attributeRankingOrderScore": 1.0, "score": 1.0 },
        "exactness": { "order": 5, "matchType": "noExactMatch", "score": 0.3333 }
      }
    }
  ]
}

Each key in _rankingScoreDetails corresponds to a ranking rule, and its score property shows how well the document performed on that rule.

How the score is computed

Ranking rules sort documents either by relevancy (words, typo, proximity, exactness, attributeRank, wordPosition) or by the value of a field (sort). Since sort does not rank documents by relevancy, it does not influence the _rankingScore.

Meilisearch computes the overall score by combining the subscores from each ranking rule, weighted by their position in the ranking rules list. Rules listed earlier carry more weight.

A document’s ranking score does not change based on the scores of other documents in the same index.

For example, if a document A has a score of 0.5 for a query term, this value remains constant no matter the score of documents B, C, or D.

Settings that influence the ranking score

The table below details all the index settings that can influence the _rankingScore. Unlisted settings do not influence the ranking score.

Index setting	Influences if	Rationale
searchableAttributes	The attributeRank ranking rule is used	The attributeRank ranking rule rates the document depending on the attribute in which the query terms show up. The order is determined by searchableAttributes
searchableAttributes	The wordPosition ranking rule is used	The wordPosition ranking rule rates the document based on the position of query terms within attributes
rankingRules	Always	The score is computed by computing the subscore of each ranking rule with a weight that depends on their order
stopWords	Always	Stop words influence the words ranking rule, which is almost always used
synonyms	Always	Synonyms influence the words ranking rule, which is almost always used
typoTolerance	The typo ranking rule is used	Used to compute the maximum number of typos for a query

Example: reading ranking score details

Consider a recipe search with two documents matching “chicken curry”, sorted by prep_time_minutes:asc:

[
  { "id": 1, "title": "Easy Chicken Curry", "prep_time_minutes": 20 },
  { "id": 2, "title": "Chicken Stew with Curry Spices and Vegetables", "prep_time_minutes": 15 }
]

With Sort placed after Proximity in ranking rules (["words", "typo", "proximity", "sort", ...]), walk through the _rankingScoreDetails in order:

Step	Rule	Doc 1	Doc 2	Outcome
0	Words	2/2, score 1.0	2/2, score 1.0	Tie
1	Typo	0 typos, score 1.0	0 typos, score 1.0	Tie
2	Proximity	score 1.0	score 0.5	Doc 1 wins

Proximity broke the tie: “chicken” and “curry” sit next to each other in Doc 1’s title (score 1.0), but are separated by three words in Doc 2 (score 0.5). Sort never got a chance to act, so even though Doc 2 has a faster prep time, it ranks second.

Notice that Sort shows a value (not a score) because it does not measure relevance. This is why a document with a higher _rankingScore can still rank lower when Sort takes priority. See ordering ranking rules for how Sort placement changes outcomes.

Use cases

• Debugging relevancy: Use showRankingScoreDetails to understand exactly why a document ranks higher or lower than expected. This helps you fine-tune ranking rules, searchable attributes, and other settings.
• Building confidence indicators: Display the ranking score in your UI as a relevancy badge or progress bar so users can gauge how closely a result matches their query.
• Setting score thresholds: Filter out low-quality results in your frontend by only displaying documents above a certain _rankingScore threshold (for example, 0.5).
• A/B testing ranking configurations: Compare ranking scores across different index configurations to measure which setup produces better relevancy for your use case.

Next steps

Built-in ranking rules

Understand the ranking rules that determine document relevancy

Custom ranking rules

Add your own ranking rules based on document attributes

Search API reference

Full reference for search parameters including showRankingScore