An OpenAI-compatible gateway to GPT, Claude, Gemini, and other models. Pay-as-you-go pricing, no subscriptions, no minimums.
Today's best model won't be the best next month. RelayModels gives you a single key to every provider: switch models with a string in your request, while we handle failover, limits, and billing.
Assistants, auto-replies, summarization
Add a chat assistant, text generation, and document parsing with one key — no separate contracts with every vendor.
from openai import OpenAI
client = OpenAI(
base_url="https://api.relaymodels.com/v1",
api_key=RELAY_KEY,
)
resp = client.chat.completions.create(
model="gpt-5-mini",
messages=[{"role": "user", "content": prompt}],
)from $0.005 per 1K tokens, one key, automatic provider failover
Text and code. How to connect
Streaming responses in OpenAI format: reasoning and function calling — all in one request from your code.
Images and files. How to send
Photos, audio, video, and PDFs go through the same request — no separate service or second key required.
Logs. See usage per request
Tokens used, cost, and which model responded — for every single call.
No subscriptions, no minimum payments. Top up your balance from 100 ₽, spend is deducted as you use it, and the exchange rate is fixed by the Central Bank of Russia at the moment of the charge.
The same key works from your backend, workers, and local tools. Limits, keys, and logs live in your dashboard.
Talk to us. Let's discuss volumes, limits, and a startup grant — up to $100 for API usage.
For businessStart building. Get a key in a minute and send your first request from a familiar SDK.
Get a key