> ## Documentation Index > Fetch the complete documentation index at: https://docs.gp.scale.com/llms.txt > Use this file to discover all available pages before exploring further. # Query Relevant Chunks > ### Description Query a knowledge base for text chunks that are most semantically relevant to the query. ### Details Given a query expressed as an embedding, this API runs a similarity search amongst the embeddings indexed in the knowledge base to find the most relevant chunk embeddings. To use this API, specify the `knowledge_base_id` of the knowledge base you want to query, pass in the natural language `query` that you want to search for, specify the value `top_k`, which is the number of similar chunks that will be returned, and specify whether you want the returned chunks to `include_embeddings`. Similarity search is used to efficiently find, retrieve, and rank chunks based on their similarity to a given query, which is also expressed as an embedding. Similarity scores ( using the cosine similarity metric) are calculated between each chunk embedding and the embedded query, and the chunks are ranked based on similarity score. The top-ranked chunks are returned as the query results. We are using the Hierarchical Navigable Small World (HNSW) algorithm to perform a k nearest neighbors search in the vector space. This algorithm returns an estimate of the best k nearest neighbors and is optimized for datasets with hundreds of thousands of vectors. You can read more about the specifics of this algorithm [here]( https://opensearch.org/docs/1.0/search-plugins/knn/approximate-knn/). #### Backwards Compatibility V2 and V1 Knowledge Bases are entirely separate and not backwards compatible. Users who have existing V1 knowledge bases will need to migrate their data to V2 knowledge bases. ## OpenAPI ````yaml https://app.stainlessapi.com/api/spec/documented/sgp/openapi.yml post /v4/knowledge-bases/{knowledge_base_id}/query openapi: 3.1.0 info: title: EGP API V4 description: >- This is the parent API for all EGP APIs. If you are looking for the EGP API, please go to https://api.egp.scale.com/docs. contact: name: Scale Generative AI Platform url: https://scale.com/genai-platform version: 0.1.0 servers: - url: https://api.egp.scale.com security: [] tags: - name: Models description: Model API. paths: /v4/knowledge-bases/{knowledge_base_id}/query: post: tags: - Knowledge Bases summary: Query Relevant Chunks description: >- ### Description Query a knowledge base for text chunks that are most semantically relevant to the query. ### Details Given a query expressed as an embedding, this API runs a similarity search amongst the embeddings indexed in the knowledge base to find the most relevant chunk embeddings. To use this API, specify the `knowledge_base_id` of the knowledge base you want to query, pass in the natural language `query` that you want to search for, specify the value `top_k`, which is the number of similar chunks that will be returned, and specify whether you want the returned chunks to `include_embeddings`. Similarity search is used to efficiently find, retrieve, and rank chunks based on their similarity to a given query, which is also expressed as an embedding. Similarity scores ( using the cosine similarity metric) are calculated between each chunk embedding and the embedded query, and the chunks are ranked based on similarity score. The top-ranked chunks are returned as the query results. We are using the Hierarchical Navigable Small World (HNSW) algorithm to perform a k nearest neighbors search in the vector space. This algorithm returns an estimate of the best k nearest neighbors and is optimized for datasets with hundreds of thousands of vectors. You can read more about the specifics of this algorithm [here]( https://opensearch.org/docs/1.0/search-plugins/knn/approximate-knn/). #### Backwards Compatibility V2 and V1 Knowledge Bases are entirely separate and not backwards compatible. Users who have existing V1 knowledge bases will need to migrate their data to V2 knowledge bases. operationId: POST-V4-/v2/knowledge-bases/{knowledge_base_id}/query parameters: - name: knowledge_base_id in: path required: true schema: type: string title: Knowledge Base Id - name: x-selected-account-id in: header required: false schema: anyOf: - type: string - type: 'null' title: Account ID Header requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/QueryKnowledgeBaseRequestV2' responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/QueryKnowledgeBaseResponseV2' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - APIKeyHeader: [] x-codeSamples: - lang: Python source: |- import os from scale_gp import SGPClient client = SGPClient( api_key=os.environ.get("SGP_API_KEY"), # This is the default and can be omitted ) response = client.knowledge_bases.query( knowledge_base_id="knowledge_base_id", query="query", top_k=1, ) print(response.request_id) - lang: Go source: "package main\n\nimport (\n\t\"context\"\n\t\"fmt\"\n\n\t\"github.com/stainless-sdks/sgp-go\"\n\t\"github.com/stainless-sdks/sgp-go/option\"\n)\n\nfunc main() {\n\tclient := sgp.NewClient(\n\t\toption.WithAPIKey(\"My API Key\"),\n\t)\n\tresponse, err := client.KnowledgeBases.Query(\n\t\tcontext.TODO(),\n\t\t\"knowledge_base_id\",\n\t\tsgp.KnowledgeBaseQueryParams{\n\t\t\tQuery: sgp.F(\"query\"),\n\t\t\tTopK: sgp.F(int64(1)),\n\t\t},\n\t)\n\tif err != nil {\n\t\tpanic(err.Error())\n\t}\n\tfmt.Printf(\"%+v\\n\", response.RequestID)\n}\n" components: schemas: QueryKnowledgeBaseRequestV2: properties: query: type: string title: Query description: >- The natural language query to be answered by referencing the data ingested into the knowledge base top_k: type: integer maximum: 10000 exclusiveMinimum: 0 title: Top K description: >- Number of chunks to return. Must be greater than 0 if specified. If not specified, all chunks will be returned. include_embeddings: type: boolean title: Include Embeddings description: Whether or not to include the embeddings for each chunk default: true metadata_filters: title: Metadata Filters description: Optional filter by metadata fields, encoded as a JSON object additionalProperties: true type: object wildcard_filters: title: Wildcard Filters description: >- Optional wildcard filter for fields. Only fields that are part of the metadata will be accepted. additionalProperties: $ref: '#/components/schemas/QueryWildcardFilterItem' type: object verbose: type: boolean title: Verbose description: Enable or disable verbose logging default: false type: object required: - query - top_k title: QueryKnowledgeBaseRequestV2 QueryKnowledgeBaseResponseV2: properties: chunks: items: $ref: '#/components/schemas/ChunkV2' type: array title: Chunks description: >- An ordered list of the k most similar chunks and their similarity scores from most to least similar started_at: title: Started At description: Timestamp at which the query was begun by the server. type: string format: date-time completed_at: title: Completed At description: Timestamp at which the query was completed by the server. type: string format: date-time request_id: title: Request Id description: query request ID for verbose logging type: string type: object required: - chunks title: QueryKnowledgeBaseResponseV2 HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError QueryWildcardFilterItem: properties: value: type: string title: Value description: Wildcard filter string case_insensitive: type: boolean title: Case Insensitive description: Check case on wildcard string default: false type: object required: - value title: QueryWildcardFilterItem ChunkV2: properties: chunk_id: type: string title: Chunk Id description: The unique ID of the chunk with embedding text: type: string title: Text description: The text associated with the chunk embedding: title: Embedding description: The vector embedding of the text associated with the chunk items: type: number type: array metadata: title: Metadata description: >- Any additional key value pairs of information stored by you on the chunk with embedding default: {} additionalProperties: true type: object user_supplied_metadata: title: User Supplied Metadata description: >- Any additional key value pairs of information returned from the custom chunking. default: {} additionalProperties: true type: object attachment_url: title: Attachment Url description: Original attachment URL from which this chunk got its data from type: string title: title: Title description: Title for this chunk, for example the file name type: string score: type: number title: Score description: >- A number between 0 and 1 representing how similar a chunk's embedding is to the query embedding. Higher numbers mean that this chunk with embedding is more similar. type: object required: - chunk_id - text - score title: ChunkV2 ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context additionalProperties: true type: object required: - loc - msg - type title: ValidationError securitySchemes: APIKeyHeader: type: apiKey in: header name: x-api-key ````