API Reference

The LimAI API provides programmatic access to document processing, classification, splitting, and data management.

Base URL

https://app.limai.io/api/v1

Authentication

All endpoints require a Bearer token:

Authorization: Bearer YOUR_API_TOKEN

Addressing a model or a deployment

Wherever a path names a {deploymentId} or an {extractionSchemaId}, three identifiers resolve: a model's own ID, the ID of a deployment that points at that model, or the model's extraction schema ID.

Address a deployment and its decisions apply — the schema version pinned to it serves the request, a paused deployment refuses new work, and the document is stamped with both. Address the model and none of them do. Schema and configuration are edited on the model either way.

Most endpoints under /api/v1/deployments/{deploymentId} echo the resolution back in two fields: modelId is the model that served the request, and viaDeploymentId is the deployment it arrived through, or null when the model was addressed directly. Each endpoint's reference page shows them where they appear.

Endpoints

Document Processing

Upload and extract structured data from documents.

• Get Upload URL — Get a presigned URL to upload a file
• Process File — Extract data synchronously
• Process File Async — Extract data asynchronously
• Process Files Async — Bulk async extraction
• Submit Corrections — Submit corrections for extracted data
• Get File Data — Get extracted data for a file
• Get Files Data — Get extracted data for multiple files
• Access Link — Get a temporary access link for a file
• Get All Data — Get all extracted data for a model or deployment

Classification

Classify documents by type.

• Get Upload URL — Get a presigned URL for classification
• Classify — Classify a document synchronously
• Classify Async — Classify a document asynchronously

Splitting

Split multi-document files into individual documents.

• Get Upload URL — Get a presigned URL for splitting
• Split — Split a document synchronously
• Split Async — Split a document asynchronously
• Get Split — Get split status and results
• Get Splits — List all splits for a splitter

Deployments

Manage models and the deployments that point at them — configuration, documents, examples, and accuracy.

• Configuration — Get or update a model's configuration
• Documents — List and manage a model's documents
• Document Detail — Get document details, pages, and content
• Document Data — Get one document's extracted data, flat and slug-keyed
• Documents Data — Get flat data for several documents in one call
• Examples — Manage extraction examples
• Corrections — View corrections and patterns
• Accuracy — Get accuracy metrics for a model or deployment
• Tables — Manage a model's tables

Projects

Manage projects and their resources.

• Schema — Removed; returns 410 Gone
• Deployments — List and create a project's models
• Buckets — List and create project buckets
• Benchmarks — List and create benchmarks

Buckets

File storage and management.

• Get Bucket — Get bucket details
• Update Bucket — Update bucket configuration
• Delete Bucket — Delete a bucket
• Upload — Get upload URL for a bucket
• Files — List and manage bucket files
• Download — Download a file from a bucket
• Send To — Send a file to a model or deployment for processing

Tables & Rows

Manage extraction tables, columns, and data rows.

• Tables — List and manage tables
• Columns — List and manage columns
• Rows — Create, list, and delete rows
• Extractions — Get extracted data by table

Validations

Check extracted documents with validation rules and custom validations.

• Validations — How rule and custom validation results appear in responses and webhooks
• Validation Scripts — List, push, update and delete custom validation scripts
• Revalidate Document — Re-evaluate rules and re-run custom validations for one document

Benchmarks

Measure extraction accuracy.

• Run Benchmark — Execute a benchmark
• Get Results — Get benchmark execution results
• Files — List benchmark files
• Cells — Get cell-level comparison data
• Metrics — Get accuracy metrics

Queue

Monitor and manage processing jobs.

• Job Status — Check job processing status
• Cancel Job — Cancel a queued job

Webhooks

Receive real-time notifications for processing events.

• Setup — Configure webhook endpoints
• Management API — Create, verify, and monitor webhook subscriptions
• Document Extracted — Event fired when extraction completes
• Document Reviewed — Event fired when data is reviewed