Guide · Updated November 2026

Use the DeepSeek API with a credit card — no Chinese phone number, no Alipay

DeepSeek-class models are among the cheapest frontier-quality models you can run. The only friction is access: the official channels want a Chinese phone number, and top-ups go through Alipay or WeChat Pay. NovaToken removes that step — one OpenAI-compatible endpoint, card payments in USD, first call in under a minute.

The short version: register with an email → buy $5 of credits with any card → point your existing OpenAI SDK code at https://api.novatoken.dev/v1. Your prompts are not stored, credits never expire.

Why people get stuck with DeepSeek in the first place

DeepSeek-class models are extremely popular with developers outside China — the models are open-weight, cheap, and strong at reasoning and code. But three things block most international users:

  1. Registration — the official platform is built around Chinese phone numbers.
  2. Payment — top-ups run through Alipay or WeChat Pay, which most developers abroad simply cannot use.
  3. Compliance friction — real-name verification is impractical if you live outside mainland China.

NovaToken is a reseller gateway built specifically for this gap: we run the models on our own infrastructure and partner inference capacity, bill you in USD with a normal card, and expose everything through the standard OpenAI API shape you already use.

How it works — three steps

  1. Create an account. Email and password only. Register here.
  2. Top up $5 (or $20 / $50). Any major credit or debit card, processed securely, charged in USD. Credits never expire.
  3. Change one line of code. Set base_url to https://api.novatoken.dev/v1 and use your new key.
# Works with the OpenAI Python SDK
from openai import OpenAI

client = OpenAI(
    api_key="sk-your-novatoken-key",
    base_url="https://api.novatoken.dev/v1"
)

resp = client.chat.completions.create(
    model="deepseek-ai/DeepSeek-V4-Flash",
    messages=[{"role": "user", "content": "Explain JWT auth in 3 bullet points"}],
    stream=True
)

What models are available

All served through the same endpoint, same key, same per-token billing:

ModelBest forPrice / 1M tokens
DeepSeek-V4-FlashReasoning, long documents (1M context)$0.30 in / $1.20 out
Qwen2.5-7B-InstructFast chat, classification, tool calling$0.10 in / $0.30 out
GLM-5.3-FlashBilingual tasks, agents$0.15 in / $0.50 out
Kimi-K2.7-CodeCode, agentic workflows (262K context)$1.20 in / $5.00 out

DeepSeek-V4-Flash also supports prompt caching: input tokens that hit the upstream cache are billed at $0.006 / 1M instead of $0.30. Nothing to configure — the discount applies whenever the provider reports a cache hit, which is typical for long system prompts, agents and coding assistants.

What it costs to get started

$5. That's roughly 25 million input tokens or 11 million output tokens on DeepSeek-V4-Flash — enough to build and test a real application. Credits never expire, there is no subscription, and you can top up again any time in $5 / $20 / $50 increments.

Buy $5 of credits Create free account

FAQ

Can I really register without a Chinese phone number?

Yes. NovaToken only needs an email address and a password. Nothing is sent to your phone, no real-name verification, no local ID required.

Which cards do you accept?

Major credit and debit cards worldwide, in USD, processed by our payment provider (Creem). PayPal-style wallets are not needed — a normal card works.

Do I keep using the official DeepSeek SDK?

You keep using the standard openai SDK family (Python, Node, Go, Rust…) — every language that speaks the OpenAI API works. Chat completions, streaming SSE, function/tool calling and JSON mode are all supported.

Do you store my prompts or train on my data?

No. We keep timestamps and token counts for billing only, purged on a rolling basis. Request bodies are not retained.

What about thinking tokens?

DeepSeek-V4-Flash is a reasoning model: it thinks before answering, and thinking tokens are billed as output (industry standard). With stream: true the first bytes still arrive quickly. If you want zero thinking delay, use Qwen2.5-7B-Instruct.

What if I run out of credits mid-project?

Top up again in $5 / $20 / $50 increments — credits are added within a minute. Nothing about your keys, code or history changes.