One endpoint · every model · pay per token

The metered gateway for LLM APIs.

TokenBeat gives you one OpenAI-compatible endpoint to every major model, billed in credits. Drop in a base URL, keep your SDK, and only pay for the tokens you use.

quickstart.py
base_url = "https://api.tokenbeat.net/v1"
# your OpenAI SDK — unchanged.
client.chat.completions.create(
  model="claude-sonnet-4",
  messages=[{"role":"user","content":"hi"}],
)
Route to every major providerAnthropicOpenAIGeminiDeepSeekGLMKimiMiniMaxQwen

Built for developers who ship

No new SDK to learn. No surprise invoices. Just tokens in, credits out.

One key, every model

Create a key, point your existing OpenAI SDK at TokenBeat, and call Anthropic, OpenAI, Google and more — switching models is a one-line change.

Transparent metering

Per-token pricing in plain credits. Real-time usage by model and key, per-key daily limits, and low-balance alerts so spend never surprises you.

Drop-in compatible

OpenAI-compatible REST and streaming. Works with the official SDKs, Claude Code, Codex and anything that speaks the chat-completions API.

    TokenBeat — The metered gateway for LLM APIs