Works with any OpenAI client

One key. Every model. Pay for what you use.

Change one base URL and reach hundreds of models. Every request is priced on its own line, so you always know what a call cost.

Prepaid credits. No subscription and no monthly minimum.
A spend limit and an expiry on every key.
Your own provider keys work too.

Get an API key Read the quickstart

Model

Your prompt

curl https://api.tiny-router.dev/v1/chat/completions \
  -H "Authorization: Bearer $TR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o-mini",
    "messages": [
      {
        "role": "user",
        "content": "Explain what a token is in one line."
      }
    ]
  }'

Input: —
Output: —
Context: —

Try it in chat

One endpoint

Text, images, audio and video

Chat, completions, responses, embeddings, image generation, speech, transcription and video jobs, all through one key and one base URL.

Browse the catalog

Speed

Streaming that stays out of the way

Responses are forwarded byte for byte and flushed on every read, so time to first token is the provider's, not ours.

Read about streaming

Cost control

Every request accounted for

Per-request tokens, latency and the price you paid, with per-key spend limits and a wallet that cannot go negative.

Limits and keys

Your accounts

Bring your own keys

Point a key at your own OpenRouter, OpenAI, Anthropic or Google account and we meter the traffic instead of reselling it.

How BYOK works

Recently added

View all

Three steps to your first call

Get an API key

Create an account
Email and password, or continue with Google. No card needed to look around.
Add credits
Top up by card or UPI through Razorpay, or send USDC on Base or Polygon from your own wallet.
Mint a key and call it
Point any OpenAI-compatible client at our base URL. Set an expiry and a spend limit per key.