0.0.11-nightly ✕ [API Error: Connection error.]

What happened?

running the qwen installation with inhouse, I get the error:
✕ [API Error: Connection error.]
stacktrace

Error: Connection error.
    at OpenAI.makeRequest (file:///usr/local/lib/node_modules/@qwen-code/qwen-code/node_modules/openai/client.mjs:265:19)
    at process.processTicksAndRejections (node:internal/process/task_queues:105:5)
    at async OpenAIContentGenerator.generateContentStream (file:///usr/local/lib/node_modules/@qwen-code/qwen-code/node_modules/@qwen-code/qwen-code-core/dist/src/core/openaiContentGenerator.js:216:29)
    at async retryWithBackoff (file:///usr/local/lib/node_modules/@qwen-code/qwen-code/node_modules/@qwen-code/qwen-code-core/dist/src/utils/retry.js:62:20)
    at async GeminiChat.makeApiCallAndProcessStream (file:///usr/local/lib/node_modules/@qwen-code/qwen-code/node_modules/@qwen-code/qwen-code-core/dist/src/core/geminiChat.js:348:32)
    at async file:///usr/local/lib/node_modules/@qwen-code/qwen-code/node_modules/@qwen-code/qwen-code-core/dist/src/core/geminiChat.js:297:40
    at async Turn.run (file:///usr/local/lib/node_modules/@qwen-code/qwen-code/node_modules/@qwen-code/qwen-code-core/dist/src/core/turn.js:47:30)
    at async GeminiClient.sendMessageStream (file:///usr/local/lib/node_modules/@qwen-code/qwen-code/node_modules/@qwen-code/qwen-code-core/dist/src/core/client.js:416:26)
    at async file:///usr/local/lib/node_modules/@qwen-code/qwen-code/dist/src/ui/hooks/useGeminiStream.js:352:26
    at async file:///usr/local/lib/node_modules/@qwen-code/qwen-code/dist/src/ui/hooks/useGeminiStream.js:448:38

What did you expect to happen?

get answer

Client information

$ qwen /about
│ CLI Version       0.0.11-nightly.1
│ Git Commit        b54fd6df
│ Model             Qwen3-Coder
│ Sandbox           no sandbox
│ OS                darwin
│ Auth Method       openai

for Release v0.0.10 - same

Login information

No response

Anything else we need to know?

No response

Reproduce the inhouse installation failure, then trace the request path through openaiContentGenerator.js, retry.js, geminiChat.js, client.js, and useGeminiStream.js as shown in the stack trace. Compare the behavior with the OpenAI authentication and connection settings; done means the request succeeds or the connection failure has a documented, actionable cause.