One gateway for
every AI capability

Connect models, images, audio, and agent tools through controlled routing, budgets, and end-to-end observability.

200 OK

POSThttps://api.cinatoken.com/v1/chat/completions

Request

curl -X POST /v1/chat/completions

{  "model": "your-model",  "messages": [    { "role": "user", "content": "Hello" }  ]}

Response

200 OK

{ "object": "chat.completion",  "choices": [{ "message": { "content": "..." } }]}

200 OK238 mscinatoken · route stable

An AI capability gateway built for builders

Flexible access, intelligent routing, controlled cost, and end-to-end observability.

/v1/chat/completions

/v1/responses

/v1/images/*

/v1/audio/*

/v1/tools/*

Unified protocols

Use one Gateway Base URL for model inference, media capabilities, and agent tools.

Chat · Responses · Images · Audio · Tools

Intelligent routing

Send each request surface into a route pool and fail over safely across upstream targets.

Priority · Weight · Failover · Sticky routing

Budgets and keys

Issue separate keys to real consumers and control cost with periodic budgets and actual usage.

Users · API keys · Periodic budgets · Three ledgers

Logs and analytics

Record request paths and budget events, then inspect usage, cost, and quality from multiple dimensions.

Request logs · Audit logs · Model/provider/user analytics

From provider to reliable traffic in four steps

Connect providers

Configure upstream endpoints and credentials

Publish models and routes

Define protocol surfaces, route pools, and failover

Issue user keys

Assign API keys and budgets by user, project, or customer

Observe every call

Continuously improve with logs, audits, and analytics

Run on your infrastructure

Cloudflare Workers + D1

Deploy the proxy and admin console on Cloudflare with Workers and D1.

View deployment docs

Docker + Postgres / MySQL

Self-host with Postgres or MySQL and keep control of the data plane and runtime.