Query Completion

Generate completion using relevant chunks as context.

When graph_name is provided, the query will leverage the knowledge graph to enhance retrieval by finding relevant entities and their connected documents.

Args: request: CompletionQueryRequest containing:

• query: Query text
• filters: Optional metadata filters
• k: Number of chunks to use as context (default: 4)
• min_score: Minimum similarity threshold (default: 0.0)
• max_tokens: Maximum tokens in completion
• temperature: Model temperature
• use_reranking: Whether to use reranking
• use_colpali: Whether to use ColPali-style embedding model
• graph_name: Optional name of the graph to use for knowledge graph-enhanced retrieval
• hop_depth: Number of relationship hops to traverse in the graph (1-3)
• include_paths: Whether to include relationship paths in the response
• prompt_overrides: Optional customizations for entity extraction, resolution, and query prompts
• folder_name: Optional folder to scope the operation to
• end_user_id: Optional end-user ID to scope the operation to
• schema: Optional schema for structured output auth: Authentication context

Returns: CompletionResponse: Generated text completion or structured output

Headers

authorization

string

Body

application/json

Request model for completion generation

query

string

required

Minimum length: 1

filters

object | null

k

integer

default:4

Required range: x > 0

min_score

number

default:0

use_reranking

boolean | null

use_colpali

boolean | null

graph_name

string | null

Name of the graph to use for knowledge graph-enhanced retrieval

hop_depth

integer | null

default:1

Number of relationship hops to traverse in the graph

Required range: 1 <= x <= 3

include_paths

boolean | null

default:false

Whether to include relationship paths in the response

folder_name

string | null

Optional folder scope for the operation

end_user_id

string | null

Optional end-user scope for the operation

max_tokens

integer | null

temperature

number | null

prompt_overrides

object | null

Optional customizations for entity extraction, resolution, and query prompts

Show child attributes

prompt_overrides.entity_extraction

object | null

Overrides for entity extraction prompts - controls how entities are identified in text during queries

Show child attributes

prompt_overrides.entity_extraction.prompt_template

string | null

Custom prompt template, MUST include both {content} and {examples} placeholders. The {content} placeholder will be replaced with the text to analyze, and {examples} will be replaced with formatted examples.

prompt_overrides.entity_extraction.examples

object[] | null

Examples of entities to extract, used to guide the LLM toward domain-specific entity types and patterns.

Example entity for guiding entity extraction.

Used to provide domain-specific examples to the LLM of what entities to extract. These examples help steer the extraction process toward entities relevant to your domain.

Show child attributes

prompt_overrides.entity_extraction.examples.label

string

required

The entity label (e.g., 'John Doe', 'Apple Inc.')

prompt_overrides.entity_extraction.examples.type

string

required

The entity type (e.g., 'PERSON', 'ORGANIZATION', 'PRODUCT')

prompt_overrides.entity_extraction.examples.properties

object | null

Optional properties of the entity (e.g., {'role': 'CEO', 'age': 42})

prompt_overrides.entity_resolution

object | null

Overrides for entity resolution prompts - controls how variant forms are grouped during queries

Show child attributes

prompt_overrides.entity_resolution.prompt_template

string | null

Custom prompt template that MUST include both {entities_str} and {examples_json} placeholders. The {entities_str} placeholder will be replaced with the extracted entities, and {examples_json} will be replaced with JSON-formatted examples of entity resolution groups.

prompt_overrides.entity_resolution.examples

object[] | null

Examples of entity resolution groups showing how variants of the same entity should be resolved to their canonical forms. This is particularly useful for domain-specific terminology, abbreviations, and naming conventions.

Example for entity resolution, showing how variants should be grouped.

Entity resolution is the process of identifying when different references (variants) in text refer to the same real-world entity. These examples help the LLM understand domain-specific patterns for resolving entities.

Show child attributes

prompt_overrides.entity_resolution.examples.canonical

string

required

The canonical (standard/preferred) form of the entity

prompt_overrides.entity_resolution.examples.variants

string[]

required

List of variant forms that should resolve to the canonical form

prompt_overrides.query

object | null

Overrides for query prompts - controls response generation style, format, and tone

Show child attributes

prompt_overrides.query.prompt_template

string | null

Custom prompt template for generating responses to queries. REQUIRED PLACEHOLDERS: {question} and {context} must be included in the template. The {question} placeholder will be replaced with the user query, and {context} will be replaced with the retrieved content. Use this to control response style, format, and tone.

schema

Schema for structured output, can be a Pydantic model or JSON schema dict

Response

200

application/json

Successful Response

Response from completion generation

completion

required

usage

object

required

Show child attributes

usage.{key}

integer

finish_reason

string | null

sources

object[]

Source information for a chunk used in completion

Show child attributes

sources.document_id

string

required

sources.chunk_number

integer

required

sources.score

number | null

metadata

object | null