Lingma
Lingma's official custom inference endpoint is an enterprise feature, not a generic personal OpenAI-compatible setting.
What is supported today
Alibaba documents enterprise inference service configuration (standard / dedicated editions) in the Lingma console under policy / inference service settings. Supported upstreams are product-specific (for example Bailian and DeepSeek), not a guarantee that any OpenAI gateway works.
Personal Lingma builds do not have a verified public path for pointing chat at https://apiclaw.biz/v1.
Recommended alternatives on APIClaw
If you need APIClaw models in an IDE or CLI, use one of these guides instead:
If you are on Lingma Enterprise
- Open the Lingma enterprise console → inference / policy configuration.
- Add a custom inference service only if your admin confirms OpenAI-compatible Chat Completions are accepted.
- If that path is approved, paste this endpoint:
text
https://apiclaw.biz/v1| Setting | Value |
|---|---|
| Endpoint | https://apiclaw.biz/v1 |
| API Key | Your APIClaw API key from API Keys |
| Model | Full model ID from the model hub |
- Confirm traffic in logs.
Third-party snippets that rewrite tongyi.apiBaseUrl for personal editions are unofficial and unsupported here.