FAQ

Frequently asked questions about the API Model Service, grouped by category: each entry gives a one-line answer first, with a "details" link to the relevant page. Questions about this documentation site itself (how to search, switch languages, etc.) are in the Site FAQ.

Account

Not receiving sign-up / verification emails?

Check your spam folder. Common providers like QQ Mail / 163 may block these emails — add the platform's sender address to your contacts and retry.

Details: Sign Up & Sign In

Forgot your password?

Click "Forgot Password?" on the sign-in page and reset it via your registered email.

Details: Sign Up & Sign In · Forgot Password

How do I change my password or bound email?

Console → "Profile" page (https://api-model.shuntianchuang.cn/profile).

Details: Sign Up & Sign In · Profile

Billing

Balance is sufficient — why does it still say "insufficient balance"?

Account balance and token quota are different things — most likely the token's own cap is used up or it has expired. Go to the API Keys page, edit the token, raise its quota, or extend its expiry.

Details: Billing & Top-up · Token Quota vs. Account Balance

Where can I see the charge details of each call?

The console's "Usage Logs" page records the time, model, token count, and cost of every call.

Details: Billing & Top-up · Checking Balance and Usage

API Calls

ErrorOne-line answerDetails
401Token missing, mistyped, or deletedError Codes & Troubleshooting
403Token disabled, or blocked by model scope / IP allowlistError Codes & Troubleshooting
429Rate limit exceeded — slow down and retry with backoffError Codes & Troubleshooting
5xxUpstream fluctuation — wait 1–2 minutes and retryError Codes & Troubleshooting
insufficient quotaZero balance — top up first; if balance is fine, it's a token quota issueBilling & Top-up

Models

Getting "model not found"?

Misspelled or wrongly capitalized model name, or the model isn't in your token's group — this site has distinct models like qwen-plus and Qwen-plus, so copy the exact name from Model Plaza.

Details: Models · Model Naming Conventions

Getting "context length exceeded"?

Your chat history exceeds the model's context window — trim the history or switch to a long-context model (e.g. kimi-k2.5, qwen-long).

Details: Models · Choosing a Model

Where do I find the model list and prices?

The console's Model Plaza is the authoritative real-time source (with filtering and search); model categories, billing methods, and naming conventions are in Models.

Clients

Any client fails to connect — run the three universal checks first:

  1. Is the URL right? https://api-model.shuntianchuang.cn (some clients require /v1 appended)
  2. Does the key have stray spaces / newlines? Copy it again from the API Keys page
  3. Does the client support streaming (SSE)? If not, turn off the "stream output" option

Guides for each client: App Integration

Specific Issues

Streaming call hangs until timeout?

Usually the client or an intermediate layer is buffering SSE: use -N with curl, stream=True in Python, for await in Node.js; disable response buffering on nginx / proxies; as a workaround, drop "stream": true.

Details: API Calls · Streaming

Claude Code can't connect or says the model doesn't exist?

The site's native Anthropic endpoint (/v1/messages) is open, but you must set ANTHROPIC_MODEL to a model on this site (Claude Code's default claude-* models are not hosted here).

Details: App Integration · Claude Code

How do I get an invoice / make a bank transfer?

For large top-ups, invoices, corporate bank transfers, and business cooperation, reach out via the email / phone / WeChat in Contact Us; see Billing & Top-up · Top-Up Troubles for the top-up platform's enterprise channel.

Contact Us