-
Notifications
You must be signed in to change notification settings - Fork 2.2k
Description
Problem Statement
On my PC model can process requests few hours. It is expected behavior. But proxy.rs, line 651 have hardcoded 300 second timeout. As result I see error in logs:
[1:01:06 PM]
DEBUG
Handling POST request to /chat/completions requiring model lookup in body
[1:01:06 PM]
DEBUG
Extracted model_id: Qwen3-Coder-30B-A3B-Instruct-Q6_K
[1:01:06 PM]
DEBUG
Found session for model_id Qwen3-Coder-30B-A3B-Instruct-Q6_K
[1:01:06 PM]
DEBUG
Adding session Authorization header
[1:01:06 PM]
DEBUG
Sending buffered body (172620 bytes)
[1:06:06 PM]
ERROR
Proxy request to model failed: error sending request for url (http://127.0.0.1:3439/chat/completions): operation timed out
Feature Idea
Allow to configure API server timeout via UI.