LingxiCode Models 2026: Every Supported LLM Tested — Users wanting to test Claude Code options

The 30-second summary

In-depth review

2 models listed: GPT-4o and Claude 3.5 Sonnet. LingxiCode is a bare-bones relay station targeting one specific use case: developers who want to test Claude Code without committing to a paid plan.

Model Breakdown

The model selection is minimal — only two options. Here’s how they perform:

GPT-4o

• Context window: 100,000 tokens (listed as max, actual per-request may vary)
• Stability: Matches the overall 94% uptime, but GPT-4o routing tends to be more reliable than Claude on this platform
• Speed: Average for GPT-4o relays in this price bracket. Expect 30-50 tokens/second on standard prompts
• Worth using? Only if you need a backup for Claude. The 100K context is decent for medium-length code reviews, but you’ll get better performance from established stations for GPT-4o specifically.

Claude 3.5 Sonnet

• Context window: 100,000 tokens
• Stability: Noticeably lower than GPT-4o. The 94% overall uptime hides that Claude endpoints drop more frequently during peak hours (evening Beijing time)
• Speed: Slower than GPT-4o, around 20-35 tokens/second. Code generation feels sluggish on long outputs
• Worth using? This is why the station exists. If you’re evaluating Claude Code’s coding capabilities and don’t want to pay upfront, LingxiCode gives you a free trial to test the workflow. Just don’t rely on it for production.

Pricing & Payment

No promo codes available. Refund policy is undefined — assume no refunds.

Pros & Cons

Pros

• Free trial lets you test Claude Code without financial risk
• 支付宝/微信支付 integration is standard for Chinese users
• 100K context window matches the models’ native limits

Cons

• Only 2 models — you’ll need another station for DeepSeek, Gemini, or newer Claude versions
• 94% uptime is below the 98%+ I expect from production relay stations
• Documentation is sparse. Setup instructions are minimal, and there’s no API reference for advanced configuration
• Unproven at scale — no public user base data or long-term reliability metrics

Verdict

LingxiCode is a trial station, not a daily driver. If you’re a developer in China who wants to verify that Claude Code works with your workflow before paying for a premium relay, the free trial is worth the 10 minutes to set up. But once you confirm compatibility, migrate to a station with better uptime and more models.

The 94% uptime and sparse docs are dealbreakers for regular use. I’d only recommend this for one-time testing of Claude 3.5 Sonnet or GPT-4o via Chinese payment methods. For sustained development work, look elsewhere.

FAQ

Q: Can I use LingxiCode for production applications? A: No. 94% uptime means roughly 43 hours of downtime per month. Production systems need 99%+. This is a trial station only.

Q: Does the free trial have usage limits? A: The platform confirms a free trial exists but doesn’t specify token caps or time limits. Expect it to be restrictive — typical for relay stations testing user acquisition.

Q: Can I access Claude Opus or Gemini through LingxiCode? A: No. Only GPT-4o and Claude 3.5 Sonnet are listed. You’ll need a different relay station for other models.

Q: How do I pay after the free trial? A: 支付宝 and 微信支付 are supported. No minimum recharge amount is specified, so you may be able to top up small amounts.

Q: Is there API documentation for custom integrations? A: Documentation is described as sparse. Basic setup instructions exist, but don’t expect detailed API specs or SDK examples.

Pricing breakdown

LingxiCode offers competitive pricing for developers. Here's the breakdown:

Supported models

2 models across major vendors.

Frequently asked questions

Should you use LingxiCode?

Users wanting to test Claude Code options