langchain_parallel

Settings for excerpt extraction.

Fetch policy for cache vs live content.

Settings for full content extraction.

Domain allow/deny lists and freshness floor for web research.

Apex-domain semantics: include nature.com, not https://www.nature.com. Wildcards permitted (e.g. .org).

Parallel Web chat model integration.

This integration connects to Parallel's Chat API, which provides real-time web research capabilities through an OpenAI-compatible interface.

Parallel Extract Tool.

Calls Parallel's Extract API to pull clean, structured content from web pages. Returns a compact summary string the LLM sees and the full structured response as a tool artifact.

One entry in the FindAll exclude_list.

One boolean condition the API uses to filter candidates.

The pair (name, description) names a True/False question the API answers per candidate; only candidates with all conditions True are returned.

Run a Parallel FindAll discovery and return the matched candidates.

FindAll discovers entities from the web that satisfy a natural-language objective plus a set of boolean match conditions. Useful for lead generation, market mapping, and entity enumeration.

Webhook config for a Parallel monitor.

Per the create-monitor API, the webhook object is {url, event_types}. The signing secret is configured at the org webhook-endpoint level (in the Parallel dashboard), not per-monitor — see :func:langchain_parallel.tasks.verify_webhook for verification.

Manage scheduled web monitors via the Parallel Monitor API (alpha).

Each method is a thin wrapper around an HTTP call. Returned dicts are the API response bodies as-is.

LangChain retriever that returns Parallel Search results as Documents.

Drops in to any RAG pipeline that expects a BaseRetriever. Each Document.page_content is the joined excerpt list from one Parallel result; metadata carries url, title, publish_date, search_id, the original excerpts list, and the query that produced it.

Parallel Search tool with web research capabilities.

This tool calls Parallel's Search API, which streamlines the traditional search -> scrape -> extract pipeline into a single API call. It supports natural-language objectives, keyword queries, domain filters, two modes (basic, advanced), location targeting, and async usage.

One BYOMCP server description for a Task Run.

Mirrors :class:parallel.types.McpServerParam so callers don't have to import from the SDK directly.

High-level Runnable for Parallel deep-research tasks.

Defaults to the pro-fast processor -- the -fast variant of pro ("Exploratory web research") at 2-5x the speed for similar accuracy. Drop the -fast suffix (processor="pro", 2-10 min) or step up to processor="ultra" (5-25 min) for the most thorough multi-source investigative reports. For shorter, enrichment-style structured tasks, use :class:ParallelEnrichment or pass processor="core-fast" here.

Always returns the full basis (citations + reasoning + confidence) on the result; lower friction than wiring up :class:ParallelTaskRunTool manually when all you want is "do deep research on this question."

Batch task runner backed by the Task Group API.

Use when you have a list of inputs and want them all processed in parallel. Returns a list of result dicts in the same order as the input list.

Run a single Parallel Task synchronously and return the structured result.

This is the agent-friendly path: an LLM calls the tool with input, the tool blocks until the Parallel Task Run completes, and returns a dict containing the output, citations (basis), and run metadata.

For long-running deep-research tasks, prefer :class:ParallelDeepResearch, which is a :class:~langchain_core.runnables.Runnable and returns the same result shape.

ParallelWeb tools.

LangChain integration for Parallel's FindAll API.

FindAll discovers entities from the web that match a natural-language objective plus a set of boolean match conditions. Returns a ranked list of candidates with citations.

LangChain retrievers backed by the Parallel Search API.

Parallel Extract Tool for LangChain.

LangChain integration for Parallel's Task API.

Four primary surfaces:

• :class:ParallelTaskRunTool — agent-callable tool that runs a single Parallel Task synchronously (via client.task_run.execute) and returns the structured result with citations.
• :class:ParallelDeepResearch — high-level :class:~langchain_core.runnables.Runnable wrapper for deep-research tasks. Defaults to the core processor and always returns the full basis (citations + reasoning + confidence).
• :class:ParallelTaskGroup — batch executor backed by the Task Group API. The general building block for fan-out/fan-in workloads.
• :class:ParallelEnrichment — the structured-batch counterpart to :class:ParallelDeepResearch. A typed Runnable that enriches a list of records against a pydantic input/output schema using the Task Group API under the hood.

Helpers: :func:build_task_spec constructs a Task Spec dict from pydantic classes; :func:verify_webhook validates HMAC signatures on incoming Parallel webhooks.

Parallel Web chat model integration.

This module provides the ChatParallelWeb class for interacting with Parallel's Chat API through an OpenAI-compatible interface.