Skip to main content

​
1. Where can I view conversation history?

The platform does not store user conversation data (for privacy protection). If you need conversation history, we recommend:
  • Using third-party clients such as ChatBox, Cherry Studio, etc.
  • Saving logs locally

​
2. Why does Prompt Caching show a hit rate of 0?

Please confirm:
  1. The model you are using supports caching (such as Claude 3.5 Sonnet, GPT-4o, etc.; GPT-5-mini is not currently supported)
  2. The preset prompt prefix for caching must be exactly the same (character-by-character match)
  3. The first call writes to the cache (incurring write costs); subsequent calls can then result in cache hits

​
3. Are the platform models proxied or self-deployed?

Closed-source models (such as OpenAI, Claude, and Gemini) are official API proxies; some open-source models are deployed by the platform itself. All models are provided through a unified gateway for stable access.
Contact Support If the FAQs above do not resolve your issue, please contact technical support through the following methods:
  • WeCom/WeChat technical support group (recommended, fastest response)
  • Information to provide:
    • Issue description + screenshots
    • Account ID (UUID)
    • Trace ID (if available, usually in the error message)
    • Request parameters (after removing sensitive information)