Skip to main content

​
What is Continum?

Continum is a compliance infrastructure that audits every LLM interaction in your application without adding latency to user responses. It provides comprehensive safety and compliance monitoring for AI applications.

Get started in 5 minutes

Install the SDK and start auditing your LLM calls

​
The Problem

Modern AI applications face critical challenges:
  • LLMs can leak PII, generate biased content, or be jailbroken
  • Manual review doesn’t scale to production traffic
  • Blocking calls for compliance checks adds 2-5 seconds of latency
  • Existing tools are either too slow or miss edge cases

​
The Solution

Continum provides zero-latency compliance auditing through asynchronous processing:
  1. SDK intercepts your LLM calls (OpenAI, Claude, Gemini)
  2. Returns response instantly to your users (0ms added latency)
  3. Mirrors the interaction asynchronously to isolated sandboxes
  4. Validates compliance using Claude as the judge
  5. Stores signals in your dashboard for monitoring and alerts
import { Continum } from '@continum/sdk';

const continum = new Continum({
  continumKey: process.env.CONTINUM_KEY,
  openaiKey: process.env.OPENAI_API_KEY
});

const response = await continum.llm.openai.gpt_4o.chat({
  messages: [{ role: 'user', content: 'Hello' }],
  sandbox: 'your-sandbox-slug'  // Specify sandbox for auditing
});
// ✅ User gets response instantly
// ✅ Compliance audit runs in background
// ✅ Violations appear in dashboard

​
Key Features

Zero Latency

Users get responses instantly while compliance runs asynchronously

Guardian Protection

Pre-LLM PII detection and redaction in under 100ms

15+ Sandbox Types

PII, bias, security, prompt injection, agent safety, and more

Real-time Dashboard

Monitor violations, track usage, and export compliance reports

​
Comprehensive Detection

Continum detects a wide range of compliance issues:

PII Detection

Names, emails, SSN, credit cards, health data

Bias Detection

Racial, gender, age, disability discrimination

Security Audit

SQL injection, XSS, secret leaks

Prompt Injection

Jailbreaks, system prompt extraction

Agent Safety

Infinite loops, irreversible actions, scope creep

Hallucination

Fake citations, false confidence

​
Architecture

Understand the architecture

Learn how Continum’s distributed architecture enables zero-latency compliance

​
Next Steps

Quickstart

Get up and running in 5 minutes

SDK Installation

Install and configure the SDK

API Reference

Explore the REST API

Dashboard Guide

Navigate the compliance dashboard