diff --git a/docs/reference/mapping/types/semantic-text.asciidoc b/docs/reference/mapping/types/semantic-text.asciidoc index ac23c153e01a3..893e2c6cff8ed 100644 --- a/docs/reference/mapping/types/semantic-text.asciidoc +++ b/docs/reference/mapping/types/semantic-text.asciidoc @@ -13,25 +13,47 @@ Long passages are <> to smaller secti The `semantic_text` field type specifies an inference endpoint identifier that will be used to generate embeddings. You can create the inference endpoint by using the <>. This field type and the <> type make it simpler to perform semantic search on your data. +If you don't specify an inference endpoint, the <> is used by default. Using `semantic_text`, you won't need to specify how to generate embeddings for your data, or how to index it. The {infer} endpoint automatically determines the embedding generation, indexing, and query to use. +If you use the ELSER service, you can set up `semantic_text` with the following API request: + [source,console] ------------------------------------------------------------ PUT my-index-000001 +{ + "mappings": { + "properties": { + "inference_field": { + "type": "semantic_text" + } + } + } +} +------------------------------------------------------------ + +NOTE: In Serverless, you must create an {infer} endpoint using the <> and reference it when setting up `semantic_text` even if you use the ELSER service. + +If you use a service other than ELSER, you must create an {infer} endpoint using the <> and reference it when setting up `semantic_text` as the following example demonstrates: + +[source,console] +------------------------------------------------------------ +PUT my-index-000002 { "mappings": { "properties": { "inference_field": { "type": "semantic_text", - "inference_id": "my-elser-endpoint" + "inference_id": "my-openai-endpoint" <1> } } } } ------------------------------------------------------------ // TEST[skip:Requires inference endpoint] +<1> The `inference_id` of the {infer} endpoint to use to generate embeddings. The recommended way to use semantic_text is by having dedicated {infer} endpoints for ingestion and search. @@ -40,7 +62,7 @@ After creating dedicated {infer} endpoints for both, you can reference them usin [source,console] ------------------------------------------------------------ -PUT my-index-000002 +PUT my-index-000003 { "mappings": { "properties": { diff --git a/docs/reference/search/search-your-data/semantic-search-semantic-text.asciidoc b/docs/reference/search/search-your-data/semantic-search-semantic-text.asciidoc index 398613ab75abb..45546d20c3327 100644 --- a/docs/reference/search/search-your-data/semantic-search-semantic-text.asciidoc +++ b/docs/reference/search/search-your-data/semantic-search-semantic-text.asciidoc @@ -21,45 +21,11 @@ This tutorial uses the <> for demonstra [[semantic-text-requirements]] ==== Requirements -To use the `semantic_text` field type, you must have an {infer} endpoint deployed in -your cluster using the <>. +This tutorial uses the <> for demonstration, which is created automatically as needed. +To use the `semantic_text` field type with an {infer} service other than ELSER, you must create an inference endpoint using the <>. -[discrete] -[[semantic-text-infer-endpoint]] -==== Create the {infer} endpoint - -Create an inference endpoint by using the <>: - -[source,console] ------------------------------------------------------------- -PUT _inference/sparse_embedding/my-elser-endpoint <1> -{ - "service": "elser", <2> - "service_settings": { - "adaptive_allocations": { <3> - "enabled": true, - "min_number_of_allocations": 3, - "max_number_of_allocations": 10 - }, - "num_threads": 1 - } -} ------------------------------------------------------------- -// TEST[skip:TBD] -<1> The task type is `sparse_embedding` in the path as the `elser` service will -be used and ELSER creates sparse vectors. The `inference_id` is -`my-elser-endpoint`. -<2> The `elser` service is used in this example. -<3> This setting enables and configures {ml-docs}/ml-nlp-auto-scale.html#nlp-model-adaptive-allocations[adaptive allocations]. -Adaptive allocations make it possible for ELSER to automatically scale up or down resources based on the current load on the process. +NOTE: In Serverless, you must create an {infer} endpoint using the <> and reference it when setting up `semantic_text` even if you use the ELSER service. -[NOTE] -==== -You might see a 502 bad gateway error in the response when using the {kib} Console. -This error usually just reflects a timeout, while the model downloads in the background. -You can check the download progress in the {ml-app} UI. -If using the Python client, you can set the `timeout` parameter to a higher value. -==== [discrete] [[semantic-text-index-mapping]] @@ -75,8 +41,7 @@ PUT semantic-embeddings "mappings": { "properties": { "content": { <1> - "type": "semantic_text", <2> - "inference_id": "my-elser-endpoint" <3> + "type": "semantic_text" <2> } } } @@ -85,9 +50,16 @@ PUT semantic-embeddings // TEST[skip:TBD] <1> The name of the field to contain the generated embeddings. <2> The field to contain the embeddings is a `semantic_text` field. -<3> The `inference_id` is the inference endpoint you created in the previous step. -It will be used to generate the embeddings based on the input text. -Every time you ingest data into the related `semantic_text` field, this endpoint will be used for creating the vector representation of the text. +Since no `inference_id` is provided, the <> is used by default. +To use a different {infer} service, you must create an {infer} endpoint first using the <> and then specify it in the `semantic_text` field mapping using the `inference_id` parameter. + + +[NOTE] +==== +If you're using web crawlers or connectors to generate indices, you have to <> for these indices to include the `semantic_text` field. +Once the mapping is updated, you'll need to run a full web crawl or a full connector sync. +This ensures that all existing documents are reprocessed and updated with the new semantic embeddings, enabling semantic search on the updated data. +==== [discrete]