Every model.
One API key.
Call models from different providers through one OpenAI-compatible endpoint. You pay each model's listed price from prepaid credit. No subscription.
ModelInputOutput
Keep your code. Change two lines.
TraiAngle speaks the OpenAI API, so the SDKs and tools you already use work as they are. Set the base URL and your TraiAngle key, then pick any model by name.
Not sure which model? Send traiangle/auto and each request goes to the cheapest model that can handle it, with automatic fallback if a provider fails.
curl https://outstanding-energy-production-1c27.up.railway.app/v1/chat/completions \
-H "Authorization: Bearer $VANOLLY_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-chat",
"messages": [{"role": "user", "content": "Hello!"}]
}'Up and running in a few minutes
- 1
Create an account
Sign up with email or Google. It takes a minute.
- 2
Add credit
Prepaid, in US dollars. Requests stop when it runs out — no surprise bills.
- 3
Call any model
One key, one endpoint. Change the model name to switch providers.
What you pay
- Each request
- The model's listed price per token
- Adding credit
- 5% on top of the amount
- Subscription
- None
- Usage
- Itemised per key and per model