One key · Every model

The token API for every model,
behind one key.

喚醒智慧的源頭雨象。

OpenAI-compatible requests, Anthropic Messages, and Gemini-style routes behind project-scoped keys, with usage logs and credits in one console.

Project-scoped model discovery$5 min top-up · credits never expire
quickstart.py · OpenAI SDK
from openai import OpenAI

client = OpenAI(
  base_url="https://dev.atptoken.ai/v1",
  api_key="sk-atp-•••"
)

r = client.chat.completions.create(
  model="<model from /v1/models>",
  messages=[{"role":"user","content":"hi"}],
)
Models

Multi-model routing

One key · Project controls
model
OpenAI-compatible
Use the same SDK shape and swap only the base URL.
model
Project-scoped
Each key follows the models enabled on its project.
model
Fallback-ready
Route through the configured provider pool for that model.
model
Discoverable
API keys can list their available models from the Gateway.
Ready to ship

Create a project key, change your base URL, and start sending requests.

$5 min top-up · failover included · per-request audit logs.