FeaturesUse CasesBlogAPI ReferenceWhy CorePlexMLPricing
Start Free
Documentation

API Reference

320+ RESTful endpoints. Cookie or Bearer token authentication. Comprehensive coverage of every CorePlexML feature.

Python SDKQuick Start
Getting Started

Authentication

CorePlexML supports two authentication methods depending on your integration pattern. API clients use Bearer tokens; browser-based sessions are handled automatically via cookies.

Bearer Token (API)

Include your API key in the Authorization header as a Bearer token. Ideal for scripts, SDKs, and server-to-server communication.

Cookie Session (UI)

Automatic for browser-based interactions. After login, an HTTP-only session cookie authenticates all subsequent requests with CSRF protection.

cURL
# List all projects
curl -X GET https://api.coreplexml.io/api/projects \
  -H "Authorization: Bearer sk_your_api_key" \
  -H "Content-Type: application/json"
Python
import requests

resp = requests.get(
    "https://api.coreplexml.io/api/projects",
    headers={"Authorization": "Bearer sk_your_api_key"}
)
Endpoints

API Modules

Eight modules covering the full ML lifecycle. Each endpoint follows consistent REST conventions with JSON payloads.

AutoML & Models

~20 endpoints

Train production-grade models with automated algorithm selection, hyperparameter tuning, and stacked ensembles.

POST/api/experiments
GET/api/experiments/{id}/status
GET/api/experiments/{id}/explain
POST/api/models/{id}/predict

MLOps & Deployments

~40 endpoints

Deploy models to production with canary rollouts, A/B testing, drift detection, and auto-retraining policies.

POST/api/mlops/projects/{id}/deployments
POST/api/mlops/deployments/{id}/predict
POST/api/mlops/deployments/{id}/promote
GET/api/mlops/deployments/{id}/drift
POST/api/mlops/ab-tests
POST/api/mlops/retraining-policies

Privacy Suite

~25 endpoints

HIPAA, GDPR, PCI-DSS, and CCPA compliance with 72+ PII detectors, anonymization transforms, and audit trails.

POST/api/privacy/policies
POST/api/privacy/sessions/{id}/detect
POST/api/privacy/sessions/{id}/transform
GET/api/privacy/pii-types

SynthGen

~8 endpoints

Generate statistically faithful synthetic datasets using CTGAN, CopulaGAN, and TVAE architectures.

POST/api/synthgen/models
POST/api/synthgen/models/{id}/generate
GET/api/synthgen/models/{id}

ML Studio

~8 endpoints

Interactive What-If analysis with scenario comparison, baseline predictions, and SHAP contributions.

POST/api/studio/sessions
POST/api/studio/sessions/{id}/scenarios
GET/api/studio/sessions/{id}/compare

Dataset Builder

~10 endpoints

Conversational ETL powered by LangGraph. Clean, transform, and export datasets through natural language.

POST/api/builder/sessions
POST/api/builder/sessions/{id}/chat
POST/api/builder/sessions/{id}/finalize

Datasets & Projects

~25 endpoints

Upload datasets in CSV, Excel, JSON, or XML. Organize work into projects with team access controls.

POST/api/datasets/upload
GET/api/datasets/{id}/quality
POST/api/projects
POST/api/projects/{id}/members

Reports

~5 endpoints

Generate comprehensive PDF or HTML reports for experiments, deployments, and data quality audits.

POST/api/reports
GET/api/reports/{id}/download
Example

Request & Response

A complete prediction request showing input payload, model output, feature contributions, and response latency.

Request → POST /api/mlops/deployments/{deployment_id}/predict
POST /api/mlops/deployments/{deployment_id}/predict
Content-Type: application/json
Authorization: Bearer sk_your_api_key

{
  "inputs": {
    "age": 35,
    "income": 75000,
    "credit_score": 720
  }
}
Response ← 200 OK
{
  "prediction": "approved",
  "probability": 0.94,
  "contributions": {
    "credit_score": 0.42,
    "income": 0.31,
    "age": 0.15
  },
  "model_id": "mod_abc123",
  "latency_ms": 23
}
Errors

Error Handling

All errors return a consistent JSON envelope with a human-readable detail message, HTTP status code, and error type identifier.

Error Response
{
  "detail": "Project not found",
  "status_code": 404,
  "error_type": "not_found"
}

Status Codes

400Bad RequestInvalid parameters or malformed JSON
401UnauthorizedMissing or invalid authentication
403ForbiddenInsufficient permissions for this resource
404Not FoundResource does not exist
422Validation ErrorRequest body failed schema validation
429Rate LimitedToo many requests, slow down
500Server ErrorInternal error, contact support
Reference

Conventions & Limits

Rate Limits

Configurable per deployment. Default burst and sustained limits protect against abuse while keeping latency low.

Request Format

All request and response bodies use JSON. File uploads use multipart/form-data for CSV, Excel, JSON, and XML.

Pagination

List endpoints support limit and offset query parameters. Default limit is 50, maximum is 200.

Async Operations

Long-running jobs (AutoML, report generation) return a job ID. Poll with GET status endpoints until completion.

File Uploads

Use multipart/form-data with the file field. Supported: CSV (.csv), Excel (.xlsx), JSON (.json), and XML (.xml).

Versioning

API is versioned at the path level. Current stable version is v1. Breaking changes are communicated 90 days in advance.

Ready to integrate?

Get your API key and start building with CorePlexML in minutes. Full SDK support for Python with type hints and async.

Get API KeyPython SDKFull Documentation