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
Preview mode lets you quickly and inexpensively test your FindAll queries with a small sample of candidates before committing to a full run. It’s ideal for validating your match conditions and enrichments.
When to use preview:
.md to its URL or sending Accept: text/markdown.- Test query structure before running on large datasets
- Validate match conditions work as expected
- Iterate quickly on FindAll schema and descriptions
How Preview Works
Preview mode uses the same API endpoint as regular FindAll runs, but withgenerator: "preview". Set match_limit from 5 to 10 to control how many candidates are evaluated. Preview results can include both matched and unmatched candidates.
Preview vs. Full Run
Key Characteristics
- Fast & Cost-Effective: Much faster and cheaper than full runs
- Sample Size: Evaluates 5–10 candidates, based on
match_limit, with no guarantee of match rate - Output Contract: Preview candidates use the same
outputandbasisfields as regular candidates; reasoning and citations are included when available - Preview Limit Semantics:
match_limitmust be between 5 and 10 and is interpreted as candidates to evaluate, not matches to find - No Extensions: Preview runs cannot be extended, but they can be enriched and cancelled while active
Preview candidates follow the same structure as full run candidates. See Candidates for details on candidate object structure and fields.
Quick Example
Best Practices
- Always Preview First: Run preview to validate match conditions before committing to full searches
- Review Both Results: Check matched and unmatched candidates to refine your query logic
- Test Enrichments Early: Validate enrichment outputs in preview before running at scale
- Examine Reasoning: Review the
basisfield to understand how matches were determined - Iterate Quickly: Use preview’s fast feedback loop to refine queries before full runs
Related Topics
- Quickstart Guide: Get started with FindAll API
- Generators and Pricing: Understand generator options and pricing
- Enrichments: Extract additional structured data for matched candidates
- Extend Runs: Increase match limits without paying new fixed costs
- Streaming Events: Receive real-time updates via Server-Sent Events
- Webhooks: Configure HTTP callbacks for run completion and matches
- Run Lifecycle: Understand run statuses and how to cancel runs
- API Reference: Complete endpoint documentation