The 30-second summary
+ What we liked
- Competitive pricing on main models
- Good availability rate
- Active development
− What we didn't
- Newer station — limited long-term data
- Customer support channels limited
In-depth review
Works without VPN from mainland China. Accepts 支付宝 and 微信支付 directly — no foreign credit card needed.
ZeroCode is a budget-tier relay station that covers the basics: GPT-4o and Claude 3.5 Sonnet. That’s it. Two models. If you need Gemini, DeepSeek, or any open-source model, this is not your station. But if you only need those two and want to test without committing cash, ZeroCode’s free tier makes it a low-risk entry point.
Pricing
ZeroCode runs a freemium model. You get a free trial with no upfront payment. No promo codes exist. Recharge minimums aren’t specified, so you can drop in small amounts via WeChat or Alipay as needed.
| Tier | Price | Notes |
|---|---|---|
| Free Trial | ¥0 | Limited usage, no time limit stated |
| Pay-as-you-go | Unlisted | Top up via 支付宝 or 微信支付 |
No refund policy is documented. Assume you burn credits as you use them.
Model Support
Only two models available:
- GPT-4o — standard OpenAI flagship
- Claude 3.5 Sonnet — Anthropic’s mid-tier reasoning model
Max context is 100,000 tokens, which covers most development tasks (code review, long document analysis, multi-turn conversations). No GPT-4 Turbo, no Claude Opus, no Gemini variants.
Performance & Reliability
Uptime sits at 94.0%. That’s below the 99%+ you’d expect from mature stations like API2D or WildCard. Expect occasional downtime — roughly 1.8 days per month offline. Latency from Beijing or Shanghai isn’t documented, but given the low price point, don’t expect premium routing.
Safety rating is 3/5 — middle of the road. Content filtering exists but isn’t aggressive. The composite score of 56.4/100 reflects the trade-off: cheap access, but limited models and reliability.
Pros & Cons
Pros
- Competitive pricing on GPT-4o and Claude 3.5 Sonnet
- Free trial available — test before paying
- Active development suggests bugs get fixed
- Native Alipay/WeChat Pay support
Cons
- Only two models — no DeepSeek, Gemini, or open-source options
- 94% uptime is weak for production use
- Customer support channels are limited
- New station — no long-term track record or refund policy
Verdict
ZeroCode is fine for one thing: cheap access to GPT-4o and Claude 3.5 Sonnet without a VPN. If you’re a student or hobbyist experimenting with LLM API calls, the free trial lets you test the waters. For any serious production work — where uptime matters or you need more than two models — look elsewhere. The 94% uptime alone disqualifies it for anything beyond personal tinkering.
FAQ
Can I use ZeroCode without a VPN from China?
Yes. The station is accessible directly from mainland China without VPN. It also supports Alipay and WeChat Pay.
What models are available on ZeroCode?
Only GPT-4o and Claude 3.5 Sonnet. No Gemini, DeepSeek, or other models.
Is there a free trial?
Yes, the free trial tier exists with no upfront payment. Usage limits are not specified.
How reliable is ZeroCode?
94% uptime. That means roughly 43 hours of downtime per month. Fine for testing, not for production.
What happens if I run out of credits?
You can top up via 支付宝 or 微信支付. Minimum recharge amount is not specified. No refunds are documented.
Pricing breakdown
ZeroCode offers competitive pricing for developers. Here's the breakdown:
| Plan | Price | Quota | Best for |
|---|---|---|---|
| Free | $0/mo | Free trial | Kicking the tires |
| Standard RECOMMENDED | Pay-as-you-go/mo | Unlimited usage | Solo devs · small teams |
| Enterprise | Custom | SLA · dedicated support | Teams & agencies |
Supported models
2 models across major vendors.
Frequently asked questions
Can I access this platform from China without a VPN?
Most relay stations are accessible from Chinese ISPs. Check our review for specific routing details.
What payment methods are accepted?
Payment options vary by platform. Some accept Alipay/WeChat Pay, others are USD/crypto only.
How does this compare to using OpenAI directly?
Relay stations add routing latency but provide access from restricted regions, unified billing, and multi-model fallback.
Is my API key safe?
Keys are encrypted at rest. Most platforms support per-project scoping and IP allow-lists.
Should you use ZeroCode?
Budget testing alongside established stations