Yield
ConceptsYield refers to streaming or incremental delivery of results. Instead of waiting for a job to complete and returning all results at once, DocLD can yield results incrementally as they are produced. This reduces perceived latency for extraction on large documents or batch processing.
How Yield Works in DocLD
| Mode | Description |
|---|---|
| Polling | Check job status; results returned when complete |
| Streaming | Results streamed as they become available (when supported) |
| Webhook | Webhook fires when job completes; results fetched via API |
For extraction, multi-page documents may produce field values incrementally. Streaming or chunked responses can deliver partial results before the full document is processed.
When Yield Matters
- Large documents — Extraction on long contracts or reports can take time; incremental results improve UX
- Batch processing — Process documents one-by-one and yield results per document instead of waiting for the entire batch
- Real-time feedback — Show progress or partial results to users while processing continues
API Behavior
The DocLD API supports both synchronous and asynchronous patterns. For long-running extraction or parsing, use jobs and webhooks. Streaming endpoints may be available for specific use cases; check the API reference for current support.
Related Concepts
Yield and streaming relate to job handling and the API. Webhooks provide completion notifications. Batch processing benefits from incremental result delivery.