GATEWAY

One endpoint, connect any router

TKX Token Router is a single OpenAI-compatible gateway. Other routers and platforms connect in with one key and one base URL — no code changes, drop into any OpenAI SDK. Text and video (Seedance) behind the same endpoint.

OpenAI-compatibleBearer keyOne base URL

Base URL

https://tkx.org/gw

Set this as your OpenAI base_url and pass your TKX key as the Bearer token. Any OpenAI SDK or tool works unchanged.

Endpoints

GET/gw/v1/models

The models your key can call.

POST/gw/v1/chat/completions

OpenAI-compatible chat completions (streaming supported).

POST/gw/v1/video/generations

Submit a Seedance video job — returns a task_id and status.{model, prompt, duration, resolution} → returns {task_id, status}.

GET/gw/v1/videos/{task_id}

Poll the job — status is queued / in_progress / completed; on completed the video URL is in the result.status ∈ queued / in_progress / completed; on completed the video URL is in the result.

Connect in 30 seconds

# OpenAI SDK — just change base_url + key
from openai import OpenAI
client = OpenAI(base_url="https://tkx.org/gw/v1", api_key="<your-tkx-key>")

# Video (Seedance)
curl https://tkx.org/gw/v1/video/generations \
  -H "Authorization: Bearer <your-tkx-key>" \
  -H "Content-Type: application/json" \
  -d '{"model":"doubao-seedance-2-0-260128","prompt":"a puppy on a sunny beach","duration":5,"resolution":"480p"}'
# → {"task_id":"...","status":"queued"}  then poll GET /gw/v1/videos/{task_id}

Get a key: partner onboarding is via BD — contactGet a key: partner onboarding is via BD — contact hi@tkx.org. Pricing and billing are arranged per partner.