Pipelines Overview
Submit deal decks, documents, and organizations to your pipelines programmatically. AlphaLens automatically ingests, extracts structured data, and runs your configured qualifying questions against each submission.
Overview
The Pipelines API lets you submit documents and organizations, track processing status, retrieve extracted data, and download original files. The typical workflow is:
-
Submit a document or organization via file upload, URL, binary data, or organization ID
-
Poll status to track ingestion and question execution progress
-
Retrieve values to access extracted and enriched data
-
Download the original document if needed
All submissions are processed asynchronously. You'll receive a document_id or pipeline_item_id immediately, then poll for completion.
Endpoints
-
Submitting Documents — Upload decks via file, URL, or binary stream
-
Adding Organizations — Add existing organizations to a pipeline by ID
-
Pipeline Items — List items, retrieve extracted data, and check processing status
-
Document Download — Get signed URLs for original files
Credit Costs
Document submissions themselves do not consume credits. Credits are charged asynchronously as the pipeline processes each item:
|
Action |
Cost |
|---|---|
|
LLM question (per question, per item) |
5 credits |
|
Visual Engine question (per question, per item) |
10 credits |
|
Web Browsing question (per question, per item) |
20 credits |
|
AlphaLens entity data access (per entity type used as context) |
2 credits |
Reading pipeline items, checking status, retrieving values, and downloading documents do not consume credits.
Rate Limiting
Document submissions are rate-limited per company, per minute. If you exceed the limit, the API returns a 429 error. The read endpoints (list, values, status, download) have separate, more generous rate limits. Contact support if you need higher limits for bulk ingestion.Overview
The Pipelines API lets you submit documents, track processing status, retrieve extracted data, and download original files. The typical workflow is:
-
Submit a document via file upload, URL, or binary data
-
Poll status to track ingestion and question execution progress
-
Retrieve values to access extracted and enriched data
-
Download the original document if needed
All submissions are processed asynchronously. You'll receive a document_id immediately, then poll for completion.
Endpoints
-
Submitting Documents — Upload decks via file, URL, or binary stream
-
Pipeline Items — List items, retrieve extracted data, and check processing status
-
Document Download — Get signed URLs for original files
Credit Costs
Document submissions themselves do not consume credits. Credits are charged asynchronously as the pipeline processes each item:
|
Action |
Cost |
|---|---|
|
LLM question (per question, per item) |
5 credits |
|
Visual Engine question (per question, per item) |
10 credits |
|
Web Browsing question (per question, per item) |
20 credits |
|
AlphaLens entity data access (per entity type used as context) |
2 credits |
Reading pipeline items, checking status, retrieving values, and downloading documents do not consume credits.
Rate Limiting
Document submissions are rate-limited per company, per minute. If you exceed the limit, the API returns a 429 error. The read endpoints (list, values, status, download) have separate, more generous rate limits. Contact support if you need higher limits for bulk ingestion.
