One API for every AI model
Buy credits once and reach OpenAI, Anthropic, Google and more with a single key. Transparent per-request pricing, automatic fallbacks.
from openai import OpenAI client = OpenAI(base_url="https://api.example.com/v1", api_key="sk-bf-…") resp = client.chat.completions.create( model="anthropic/claude-sonnet-4-5", messages=[{"role": "user", "content": "Hello!"}], extra_body={"fallbacks": ["openai/gpt-4o-mini"]}, ) # resp.usage.cost -> 0.000601
Everything you need to ship on many models
OpenAI-compatible endpoints for chat, responses, embeddings and more. Switch models by changing a string.
Pass a list of models. If a provider is down or rate-limited, the next one answers — you only pay for the one that did.
Top up by card, crypto or local bank card. Provider prices are passed through at cost; a small fee applies only when you buy credits.
Each response carries usage.cost. Look up any generation later, export activity as CSV, and see spend per key and model.
Per-key spending limits with daily, weekly or monthly resets, expiry dates, model allow-lists and one-click rotation.
Organizations with owners, admins and members, a shared balance, invitations and an audit trail.
Pay what the model costs. Nothing more per request.
Token prices match the providers' list prices. We charge a small platform fee when you add credits — that's it.Unused credits from a card purchase can be refunded within 24 hours.