Implementation:Langchain ai Langchain Perplexity Types

Knowledge Sources	Langchain_ai_Langchain
Domains	LLM, Perplexity, Data Models
Last Updated	2026-02-11 00:00 GMT

Overview

Pydantic data models for Perplexity API configuration options including web search settings, user location, and media response overrides.

Description

This module defines Pydantic BaseModel classes used to configure Perplexity API requests in the langchain-perplexity partner package. It includes four models: UserLocation for geographic coordinates and location metadata, WebSearchOptions for controlling search context size, search type, and image relevance, MediaResponseOverrides for toggling video and image returns, and MediaResponse as a wrapper around the overrides.

Usage

Import these types when constructing Perplexity chat model requests that require fine-grained control over web search behavior, user location context, or media response settings.

Code Reference

Source Location

• Repository: Langchain_ai_Langchain
• File: libs/partners/perplexity/langchain_perplexity/types.py
• Lines: 1-29

Signature

class UserLocation(BaseModel):
    latitude: float | None = None
    longitude: float | None = None
    country: str | None = None
    region: str | None = None
    city: str | None = None

class WebSearchOptions(BaseModel):
    search_context_size: Literal["low", "medium", "high"] | None = None
    user_location: UserLocation | None = None
    search_type: Literal["fast", "pro", "auto"] | None = None
    image_search_relevance_enhanced: bool | None = None

class MediaResponseOverrides(BaseModel):
    return_videos: bool | None = None
    return_images: bool | None = None

class MediaResponse(BaseModel):
    overrides: MediaResponseOverrides | None = None

Import

from langchain_perplexity.types import (
    UserLocation,
    WebSearchOptions,
    MediaResponseOverrides,
    MediaResponse,
)

I/O Contract

UserLocation Fields

Name	Type	Required	Description
latitude	None	No	Geographic latitude coordinate.
longitude	None	No	Geographic longitude coordinate.
country	None	No	Country name or code.
region	None	No	Region or state name.
city	None	No	City name.

WebSearchOptions Fields

Name	Type	Required	Description
search_context_size	None	No	Controls the amount of search context retrieved.
user_location	None	No	User location for geo-contextualized search results.
search_type	None	No	Search mode: fast, pro, or auto-detect.
image_search_relevance_enhanced	None	No	Whether to enhance image search relevance.

MediaResponseOverrides Fields

Name	Type	Required	Description
return_videos	None	No	Whether to include videos in the response.
return_images	None	No	Whether to include images in the response.

MediaResponse Fields

Name	Type	Required	Description
overrides	None	No	Media response override settings.

Usage Examples

Basic Usage

from langchain_perplexity.types import WebSearchOptions, UserLocation

# Configure web search options with user location
location = UserLocation(
    latitude=37.7749,
    longitude=-122.4194,
    city="San Francisco",
    region="California",
    country="US",
)

search_options = WebSearchOptions(
    search_context_size="high",
    user_location=location,
    search_type="pro",
)

Media Response Configuration

from langchain_perplexity.types import MediaResponse, MediaResponseOverrides

media = MediaResponse(
    overrides=MediaResponseOverrides(
        return_images=True,
        return_videos=False,
    )
)

Related Pages

• Langchain_ai_Langchain_Perplexity_Output_Parsers
• Langchain_ai_Langchain_PerplexityRetriever
• Langchain_ai_Langchain_PerplexitySearchTool