[BUG] <title>[BUG] Qwen Companion: Frequent "Request timed out" errors with Qwen-3.8-Max-Preview (Singapore API/Token Plan))

[nikitatwo404-prog]

是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this?

• 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions

该问题是否在FAQ中有解答？ | Is there an existing answer for this in FAQ?

• 我已经搜索过FAQ | I have searched FAQ

当前行为 | Current Behavior

When sending prompts or commands in Qwen Companion using Qwen-3.8-Max-Preview via the Singapore API endpoint (Alibaba Cloud Model Studio - Singapore), the extension hangs during processing and fails with:
"Request timed out. This may be due to a network issue. Please try again."

期望行为 | Expected Behavior

The API connection should remain active during long-context pre-fill phases (TTFT) and stream responses back reliably without triggering gateway timeouts.

复现方法 | Steps To Reproduce

1. Set up Qwen Companion extension using API Key from Alibaba Cloud Model Studio (Singapore).
2. Select model Qwen-3.8-Max-Preview.
3. Send a request with codebase context or run /stuck command.
4. Wait during the initial pre-fill phase (~30–60 seconds)
5. Receive error: "Request timed out. This may be due to a network issue. Please try again."

运行环境 | Environment

- OS: Windows 11 (x64)
- Client: Qwen Companion Extension (AntiGravity / VS Code)
- API Provider: Alibaba Cloud Model Studio (Singapore Endpoint)
- Model: Qwen-3.8-Max-Preview
- Plan: Token Plan (Individual)

备注 | Anything else?

Suspecting a strict HTTP/SSE timeout on the Singapore API gateway when handling heavy prompts / high TTFT load, combined with a lack of keep-alive mechanisms in the extension.