Deployed on Cloudflare Edge • Sub-15ms Responses

Stop burning money on LLM calls.
Cache prompts. Cap runaway bills.

The drop-in AI proxy gateway for modern engineering teams. Cut model costs by up to 40% with zero-latency edge caching and instant budget kill-switches.

Start Free (50k Requests) View 1-Line Setup

app.ts

// Just change your baseURL to activate EdgeGuard
const openai = new OpenAI({
  baseURL: "https://gateway.edgeguard.ai/v1",
  apiKey: process.env.EDGEGUARD_KEY,
  defaultHeaders: { "X-Provider-Key": process.env.OPENAI_API_KEY }
});

Engineered for Reliability and Speed

Built natively on Cloudflare's 300+ edge locations to safeguard every LLM request.

Edge KV Semantic Caching

Identical or frequent prompts are served straight from Cloudflare edge caches in 10-15ms, eliminating model API fees entirely.

Hard Spend Limits

Prevent $10,000 accidental loops. Set hard monthly budget thresholds per key. The gateway rejects excess calls before they hit your wallet.

Real-Time Cost Attribution

Know exactly which user, feature, or script costs the most. Track prompt tokens, completion tokens, and latency live in your dashboard.

Simple, Predictable Pricing

Start free, upgrade as your AI traffic scales.

Developer

For pet projects and initial staging.

$0 / forever

50,000 requests / month
1 Gateway API Key
24-hour log retention
Basic edge caching

Get Started Free

Pro Scale

For production applications in market.

$16 / month (billed annually)

1,000,000 requests / month
Unlimited API Keys
30-day analytics & logs
Hard budget kill-switches
Sub-15ms KV caching

Upgrade to Pro

Startup Team

For growing engineering organizations.

$66 / month (billed annually)

10,000,000 requests / month
90-day telemetry retention
Team member seats & roles
Slack & Discord webhook alerts
Dedicated edge worker routing

Choose Team Plan

Stop burning money on LLM calls. Cache prompts. Cap runaway bills.