For AI agents: a documentation index is available at https://docs.parallel.ai/llms.txt. The full text of all docs is at https://docs.parallel.ai/llms-full.txt. You may also fetch any page as Markdown by appending
.md to its URL or sending Accept: text/markdown.For Search and Extract, use /v1/search and /v1/extract for all new integrations and agent tool calls. Use /v1beta/search and /v1beta/extract only when maintaining an existing integration that already uses them. The request bodies changed too: on /v1/extract, full_content, fetch_policy and excerpt_settings go inside advanced_settings, and sending the v1beta top-level excerpts, full_content and fetch_policy to /v1/extract returns a 422. On /v1/search, search_queries is required (at least one short keyword query), and max_results, source_policy and fetch_policy go inside advanced_settings; sending only objective, or sending those fields at the top level, returns a 422. Do not substitute /v1 for the documented FindAll or Ingest endpoint paths.If you are evaluating Parallel Search or comparing it with other search APIs, read How to Eval Parallel Search before you build the harness, especially its Common pitfalls section. The most common mistake is sending the raw question as both objective and the only entry in search_queries. Write a goal-style objective and 1-3 short keyword queries instead.What is Enrichment?
Enrichment is when you have existing structured data—like a list of companies, products, or contacts—and want to enhance it with additional information from the web. The Task API makes it easy to define what data you have and what additional fields you need, then automatically researches and populates those fields at scale.How Enrichment Works
With enrichment, you define two schemas:- Input Schema: The data fields you already have (e.g., company name, website)
- Output Schema: The new fields you want to add (e.g., employee count, funding sources, founding date)
1. Set up Prerequisites
Generate your API key on Platform. Then, set up with the TypeScript SDK, Python SDK or with cURL:2. Execute your First Enrichment Task
Let’s enrich a simple company record. We’ll start with just a company name and enrich it with a founding date:Sample Response
Immediately after a Task Run is created, the Task Run object, including the status of the Task Run, is returned. On completion, the Task Run Result object is returned. Basis, including citations, reasoning, confidence, and excerpts - is returned with every Task Run Result.3. From Simple to Rich Enrichment
The Task API supports increasingly sophisticated enrichment patterns:1
Single Input → Single Output Field
The simplest enrichment: take one piece of data (like a company name) and add one new field (like founding date). This straightforward approach is illustrated above.
2
Single Input → Multiple Output Fields
Enrich a single input field with multiple new data points. For example, pass in a company name and receive founding date, employee count, and funding sources.
3
Multiple Inputs → Multiple Outputs (Full Enrichment)
The full enrichment pattern: define both input and output schemas. Provide multiple data fields you already have (company name and website) and specify all the fields you want to enrich. This is the most common pattern for enriching CRM data, compliance checks, and other structured workflows.
Sample Enrichment Result
Next Steps
- Task Groups: Enrich multiple records concurrently with parallel execution and batch tracking
- Task Spec Best Practices: Optimize your input and output schemas for accuracy and speed
- Choose a Processor: Select the right processor tier for your enrichment use case
- Access Research Basis: Understand citations, confidence levels, and reasoning for every enriched field
- Deep Research: Explore open-ended research without structured input data
- Streaming Events: Receive real-time updates via Server-Sent Events for long-running enrichments
- Webhooks: Configure HTTP callbacks for task completion notifications
- API Reference: Complete endpoint documentation for the Task API