Skip to content

VectorConfig

Properties

Name Type Description Notes
name String Unique name for this vector configuration
r#type Type Type of vector (dense_model, sparse_model, full_text, trigrams, whitespace, wmtr, dense_custom, sparse_custom) (enum: dense_model, sparse_model, full_text, trigrams, whitespace, wmtr, keyword, dense_custom, sparse_custom, noop)
model Option<String> Model name for transformer-based vectors (max 210 characters, e.g., 'sentence-transformers/all-MiniLM-L6-v2'). Used for document indexing. [optional]
revision Option<String> Optional model revision (max 210 characters, branch/tag/commit) for specific model version. Used for document indexing. [optional]
query_model Option<String> Optional model name for query vectorization (max 210 characters). If not specified, uses 'model' for both documents and queries. [optional]
query_revision Option<String> Optional model revision for query vectorization (max 210 characters). If not specified, uses 'revision' for both documents and queries. [optional]
dimensions Option<i32> Dimensions for the vector. Required for dense_custom vectors. For dense_model vectors, auto-detected if not specified. [optional]
top_k Option<i32> Number of top-scoring terms to keep for sparse vectors. Used by sparse_model, full_text, trigrams, whitespace, wmtr, and sparse_custom vectors. Ignored by dense vectors. [optional][default to 128]
wmtr_word_ratio Option<i32> Percentage of WMTR top_k allocated to word weights. [optional][default to 80]
index_fields Option<Vec> List of fields to index with this vector (name, description, content). Defaults to ['content'] if not specified. (enum: name, description, content) [optional]
language_default_code Option<String> Two-letter ISO 639-1 language code for language-based vector types (e.g., 'en', 'es', 'fr') [optional][default to en]
language_detect Option<bool> Whether to automatically detect language for language-based vector types [optional][default to false]
language_confidence Option<f64> Minimum confidence threshold for language detection. If detection confidence is below this value, language_default_code will be used instead. [optional][default to 0.9]
normalization Option<bool> Whether to normalize vectors. Only supported for dense vectors. Sparse vectors do not support normalization. [optional]
dense_distance Option<DenseDistance> Distance metric for dense vectors (cosine, dot, euclid). Defaults to cosine. (enum: cosine, dot, euclid) [optional][default to Cosine]
keep_case Option<bool> Whether to keep original case for text preprocessing. Only applies to model-based vectors (dense_model, sparse_model). Defaults to False (lowercase). [optional][default to false]

[Back to Model list] [Back to API list] [Back to README]