Implementation:Cohere ai Cohere python ChatSearchResult Model

Knowledge Sources	Cohere Python SDK
Domains	SDK, Search, RAG
Last Updated	2026-02-15 14:00 GMT

Overview

ChatSearchResult is a Pydantic model representing a single search result returned from a connector during retrieval-augmented generation (RAG) in the Cohere chat API.

Description

The ChatSearchResult model captures the outcome of a document search performed by a connector. It includes the search query that was issued, the connector that produced the result, the list of document IDs that were found, and optional error handling information. This model is part of the RAG pipeline where the Cohere API queries external data sources to ground generated responses with real documents.

Key characteristics:

• Contains a reference to the originating search_query via ChatSearchQuery.
• Contains a reference to the connector that performed the search via ChatSearchResultConnector.
• Returns a list of document_ids that matched the query.
• Provides an optional error_message if the search failed.
• Includes a continue_on_failure flag to control whether the chat request should proceed even if this connector fails.

Usage

Use ChatSearchResult when processing search results from the Cohere chat API's RAG pipeline. This model appears in chat responses that include connector-based document retrieval, allowing you to inspect which documents were found, which connector was used, and whether any errors occurred during the search.

Code Reference

Source Location

• Repository: Cohere Python SDK
• File: src/cohere/types/chat_search_result.py

Signature

class ChatSearchResult(UncheckedBaseModel):
    search_query: typing.Optional[ChatSearchQuery] = None
    connector: ChatSearchResultConnector = pydantic.Field()
    document_ids: typing.List[str] = pydantic.Field()
    error_message: typing.Optional[str] = pydantic.Field(default=None)
    continue_on_failure: typing.Optional[bool] = pydantic.Field(default=None)

Import

from cohere.types import ChatSearchResult

I/O Contract

Fields

Field	Type	Required	Description
search_query	Optional[ChatSearchQuery]	No	The search query that was used to retrieve documents.
connector	ChatSearchResultConnector	Yes	The connector from which this result comes from.
document_ids	List[str]	Yes	Identifiers of documents found by this search query.
error_message	Optional[str]	No	An error message if the search failed.
continue_on_failure	Optional[bool]	No	Whether a chat request should continue or not if the request to this connector fails.

Usage Examples

from cohere.types import ChatSearchResult

# ChatSearchResult is typically returned as part of a chat response
response = client.chat(
    message="What are the latest quarterly results?",
    connectors=[{"id": "web-search"}],
)

# Access search results from the response
for result in response.search_results:
    print(f"Connector: {result.connector}")
    print(f"Documents found: {result.document_ids}")
    if result.error_message:
        print(f"Error: {result.error_message}")
    if result.search_query:
        print(f"Query used: {result.search_query}")

Related Pages

• Environment:Cohere_ai_Cohere_python_Python_SDK_Runtime