Model Router — AI API Gateway

Quickstart

1

Register for an API key

Free to create. Keys are prefixed mr_sk_. Save it — it's shown once.

# Register (no auth required)
curl -X POST https://api.lxg2it.com/v1/auth/register \
  -H "Content-Type: application/json" \
  -d '{"name": "my-key"}'

2

Make a request

Drop-in replacement for the OpenAI API. Point your existing clients here.

curl -X POST https://api.lxg2it.com/v1/chat/completions \
  -H "Authorization: Bearer mr_sk_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

3

Add credits to continue

Pay-as-you-go via card. Requests are billed at cost with no markup — you pay what the upstream provider charges.

Open billing dashboard →

Endpoints

POST

/v1/auth/register

Create a new API key. Returns key and key_id.

POST

/v1/chat/completions auth

OpenAI-compatible chat completions. Supports streaming, model: "auto", and all tier aliases.

GET

/v1/models auth

List available models and tier aliases in OpenAI format.

GET

/v1/usage auth

Usage summary for your key — tokens, requests, estimated cost.

GET

/v1/billing/status auth

Credit balance and payment method status.

GET

/health

Provider health and circuit breaker status. No auth required.

Model Tiers

Use model: "auto" to let the router pick based on prompt characteristics, or specify a tier directly. Explicit model names (e.g. claude-3-5-sonnet-20241022) are passed through unchanged.

fast

claude-haiku-4
gpt-4o-mini
gemini-2.0-flash

balanced

claude-sonnet-4-5
gpt-4o
gemini-1.5-pro

capable

claude-opus-4-5
gpt-4-turbo
gemini-1.5-pro-exp

Circuit breakers automatically reroute around provider outages. Failing providers are retried after a cooldown period.