Chinese model APIs on one OpenAI-compatible key. Browse models

North America · startup pilots

No Chinese phone required

Ship a Chinese-model pilot without CN-phone friction

North American teams usually care about signup speed, USD prepaid control and an honest Live catalog. Start on Live DeepSeek routes, keep OpenAI SDK shape, and only expand after your own prompt suite passes.

EnglishSpanish (US/LATAM product surfaces)

North America pilot plan

A region-specific rollout, not a generic template

StepActionSuccess signal
Day 0Google signup → create a budgeted staging key → call deepseek-chat with 10 golden prompts.All requests return; spend stays under the key cap.
Day 1–2A/B the same prompts on deepseek-v4-pro for hard reasoning/code cases.Record quality win rate and cost per successful task.
Day 3+Promote only Live IDs to a production key with tighter budgets.Integrating catalog entries stay out of prod traffic.

Workload matrix

Where to start for North America

Coding copilots

deepseek-chat

Lower cost for high-volume edits; escalate failures to V4 Pro.

Hard reasoning / long docs

deepseek-v4-pro

Better multi-step recovery on the same rubric.

Support drafts

deepseek-chat

Fast first-pass replies with budgeted keys per queue.

Live prices

Only Live DeepSeek routes

ModelBillable price
DeepSeek Flash

deepseek-chat

$0.3 input / $1.2 output per 1M
DeepSeek V4 Pro

deepseek-v4-pro

$1.32 input / $3.96 output per 1M

Integrating catalog families are market context only. Confirm Live status on /models before production.

Replace a blocked vendor signup

If a direct Chinese vendor still asks for a mainland phone, keep your OpenAI client, point baseURL at SmarToken, and validate Live DeepSeek IDs before rewriting product code.

Finance-friendly pilots

Prepaid USD wallet plus daily/monthly key caps make experimental spend visible to finance without a separate PO for every trial.

OpenRouter migration slice

Move only Chinese text routes first. Confirm Live status on /models; do not assume every OpenRouter alias maps 1:1.

Migration

Keep the SDK, change the base URL

  • Create separate staging and production keys with different budgets.
  • Run the same 10–20 golden prompts on deepseek-chat and deepseek-v4-pro.
  • Track cost per successful task, including retries.
  • Keep Integrating catalog families out of production until they show Live.
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.SMTOKEN_API_KEY,
  baseURL: "https://thesmartoken.com/v1",
});

const completion = await client.chat.completions.create({
  model: "deepseek-chat",
  messages: [
    {
      role: "user",
      content: "Draft a English reply for our North America support pilot.",
    },
  ],
});

FAQ for North America

Do we need a Chinese phone number?

No. Overseas developers can sign in with Google and create SmarToken API keys without mainland phone verification.

Which models are Live for North America traffic?

The Live text routes are deepseek-chat and deepseek-v4-pro. Re-check /models before citing availability.

Is the API endpoint regional?

No. Use https://thesmartoken.com/v1 worldwide. This page is a workload guide, not a separate cluster.

How does this compare to OpenRouter?

Use SmarToken when you want Chinese-model routing with prepaid USD controls and explicit Live vs Integrating status. See /openrouter-alternative for the decision frame.