Every model. One API.
One key, one base URL — call OpenAI, Anthropic, Gemini, DeepSeek and more. Smart routing with automatic failover, multi-currency (CNY / USD / USDT) top-ups, transparent metered billing.
from openai import OpenAI
client = OpenAI(base_url="https://api.routeall.ai/v1", api_key="sk-ra-...")
client.chat.completions.create(model="deepseek-chat",
messages=[{"role":"user","content":"hi"}])Why RouteAll
Swap base_url and go
One key + one base URL for every model; migrate existing OpenAI SDK code with zero changes — switch models by name only.
Model marketplace
Every major provider in one catalog, with live prices for your tier and filters by capability and price.
Playground
Try any model right in the browser: multi-session chat, image input, parameter tuning — on the real billing pipeline.
Native multi-format endpoints
OpenAI /v1, Anthropic /v1/messages and Gemini /v1beta — keep whichever SDK you already use.
Smart routing & failover
Channels picked by priority/weight/health with automatic failover on upstream errors — never double-billed.
Transparent billing
Per-token and per-call metering, prompt-cache discounts, balance pre-authorization — never silently free.
Model marketplace
Supported models and public pricing (standard tier). Log in to see your tier’s pricing.
coming soon…
Pay as you go, tiered discounts
Official retail pricing per token; different user groups get different rate multipliers. Live price table in the API docs.
API DocsSubscription plans
Coming soon — Pro / Max monthly plans (bundled credits + lower multiplier)