issue: 多问题汇总

[zopenb]

Check Existing Issues

• I have searched for any existing and/or related issues.
• I have searched for any existing and/or related discussions.
• I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!).
• I am using the latest version of Open WebUI.

Installation Method

Git Clone

Open WebUI Version

v0.7.3-7

Ollama Version (if applicable)

No response

Operating System

Centos7.6

Browser (if applicable)

Chrome 141

Confirmation

• I have read and followed all instructions in README.md.
• I am using the latest version of both Open WebUI and Ollama.
• I have included the browser console logs.
• I have included the Docker container logs.
• I have provided every relevant configuration, setting, and environment variable used in my setup.
• I have clearly listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc).
• I have documented step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation. My steps:
• Start with the initial platform/version/OS and dependencies used,
• Specify exact install/launch/configure commands,
• List URLs visited, user input (incl. example values/emails/passwords if needed),
• Describe all options and toggles enabled or changed,
• Include any files or environmental changes,
• Identify the expected and actual result at each stage,
• Ensure any reasonably skilled user can follow and hit the same issue.

Expected Behavior

佬友你好，近两天在不断使用该项目，发现一些问题和建议，特来反馈。如下

模型思考问题：

1. 使用Qwen-Max模型，后端均为同一个NewAPI，接入2API chat.qwen.ai，该项目版本的OpenWebUI输出思考，但其他项目正常不输出思考，但原版的OpenWebUI在使用该模型时，也不输出思考，主要通过Modelname来区分是否输出思考，而且关闭推理强度依然会输出思考。

从后端日志看，"enable_thinking": true是开启的。

2026-01-26 02:26:44.754 | INFO | open_webui.main:process_chat:1748 - [DEBUG PROCESS_CHAT] Starting process_chat for model: qwen3-max-2025-10-30 2026-01-26 02:26:44.754 | INFO | open_webui.utils.middleware:process_chat_payload:1456 - [DEBUG PAYLOAD] process_chat_payload called for model: qwen3-max-2025-10-30, owned_by: openai 2026-01-26 02:26:44.769 | INFO | open_webui.main:process_chat:1753 - [DEBUG PROCESS_CHAT] Calling chat_completion_handler 2026-01-26 02:26:44.769 | INFO | open_webui.utils.chat:generate_chat_completion:275 - Model ID: qwen3-max-2025-10-30, Owned by: openai 2026-01-26 02:26:44.769 | INFO | open_webui.utils.chat:generate_chat_completion:305 - Routing to OpenAI handler - Model ID: qwen3-max-2025-10-30, owned_by: openai, model keys: ['id', 'name', 'owned_by', 'openai', 'urlIdx', 'connection_type', 'info', 'actions', 'filters', 'tags'] 2026-01-26 02:26:44.771 | INFO | open_webui.routers.openai:get_all_models:2075 - get_all_models() 2026-01-26 02:26:44.772 | INFO | open_webui.routers.openai:generate_chat_completion:2914 - [DEBUG] stream_options in payload: {'include_usage': True} 2026-01-26 02:26:44.773 | INFO | open_webui.routers.openai:generate_chat_completion:3024 - [DEBUG] Thinking disabled (budget=0) 2026-01-26 02:26:44.773 | INFO | open_webui.routers.openai:generate_chat_completion:3058 - [DEBUG] Full payload being sent: {"stream": true, "model": "qwen3-max-2025-10-30", "messages": [{"role": "system", "content": "User Context:\n\n"}, {"role": "user", "content": "9.8\u548c9.11\u54ea\u4e2a\u5927\uff1f"}], "stream_options": {"include_usage": true, "include_reasoning": true, "include_thinking": true}, "enable_thinking": true, "include_reasoning": true, "return_reasoning": true, "show_thinking": true}... 2026-01-26 02:26:44.775 | INFO | uvicorn.protocols.http.httptools_impl:send:483 - 192.168.0.1:0 - "POST /api/chat/completions HTTP/1.1" 200 2026-01-26 02:26:44.797 | INFO | uvicorn.protocols.http.httptools_impl:send:483 - 192.168.0.1:0 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 2026-01-26 02:26:47.850 | INFO | open_webui.routers.openai:generate_chat_completion:3083 - [DEBUG RESPONSE] r.status=200, Content-Type=text/event-stream 2026-01-26 02:26:47.850 | INFO | open_webui.routers.openai:generate_chat_completion:3344 - [DEBUG] Response status=200, Content-Type=text/event-stream, stream_requested=True 2026-01-26 02:26:47.850 | INFO | open_webui.routers.openai:generate_chat_completion:3370 - [DEBUG STREAM RETRY] compat_stream_retry=True, compat_steps=['compact_thinking', 'drop_thinking', 'drop_stream_options', 'drop_tools'] 2026-01-26 02:26:48.021 | INFO | open_webui.routers.openai:generate_chat_completion:3375 - [DEBUG STREAM RETRY] error_msg=None 2026-01-26 02:26:48.021 | INFO | open_webui.routers.openai:generate_chat_completion:3443 - [DEBUG] Streaming response detected, use_responses_api=False 2026-01-26 02:26:48.021 | INFO | open_webui.main:process_chat:1755 - [DEBUG PROCESS_CHAT] chat_completion_handler returned, response type: <class 'starlette.responses.StreamingResponse'> 2026-01-26 02:26:48.034 | INFO | open_webui.routers.openai:combined_stream:3475 - [STREAM DEBUG] Buffered line #0: b'data: {"id":"chatcmpl-54cba5b4-6db9-4226-9747-c16dd9cb3b44","object":"chat.completion.chunk","created":1769365607848,"choices":[{"index":0,"delta":{"content":"<think>\\n\\nThe"},"finish_reason":null}]}\n' 2026-01-26 02:26:48.034 | INFO | open_webui.routers.openai:combined_stream:3475 - [STREAM DEBUG] Buffered line #1: b'\n' 2026-01-26 02:26:48.034 | INFO | open_webui.routers.openai:combined_stream:3475 - [STREAM DEBUG] Buffered line #2: b'data: {"id":"chatcmpl-54cba5b4-6db9-4226-9747-c16dd9cb3b44","object":"chat.completion.chunk","created":1769365607888,"choices":[{"index":0,"delta":{"content":" user is asking"},"finish_reason":null}]}\n' 2026-01-26 02:26:48.035 | INFO | open_webui.routers.openai:combined_stream:3475 - [STREAM DEBUG] Buffered line #3: b'\n' 2026-01-26 02:26:48.035 | INFO | open_webui.routers.openai:combined_stream:3475 - [STREAM DEBUG] Buffered line #4: b'data: {"id":"chatcmpl-54cba5b4-6db9-4226-9747-c16dd9cb3b44","object":"chat.completion.chunk","created":1769365608020,"choices":[{"index":0,"delta":{"content":" which number"},"finish_reason":null}]}\n' 2026-01-26 02:26:48.035 | INFO | open_webui.routers.openai:combined_stream:3475 - [STREAM DEBUG] Buffered line #5: b'\n' 2026-01-26 02:26:48.035 | INFO | open_webui.routers.openai:combined_stream:3475 - [STREAM DEBUG] Buffered line #6: b'data: {"id":"chatcmpl-54cba5b4-6db9-4226-9747-c16dd9cb3b44","object":"chat.completion.chunk","created":1769365608020,"choices":[{"index":0,"delta":{"content":" is larger between"},"finish_reason":null}]}\n' 2026-01-26 02:26:48.035 | INFO | open_webui.routers.openai:combined_stream:3475 - [STREAM DEBUG] Buffered line #7: b'\n' 2026-01-26 02:26:48.072 | INFO | open_webui.routers.openai:combined_stream:3484 - [STREAM DEBUG] Line #1: b'data: {"id":"chatcmpl-54cba5b4-6db9-4226-9747-c16dd9cb3b44","object":"chat.completion.chunk","created":1769365608072,"choices":[{"index":0,"delta":{"content":" 9.8 and "},"finish_reason":null}]}\n' 2026-01-26 02:26:48.073 | INFO | open_webui.routers.openai:combined_stream:3484 - [STREAM DEBUG] Line #2: b'\n' 2026-01-26 02:26:48.133 | INFO | open_webui.routers.openai:combined_stream:3484 - [STREAM DEBUG] Line #3: b'data: {"id":"chatcmpl-54cba5b4-6db9-4226-9747-c16dd9cb3b44","object":"chat.completion.chunk","created":1769365608133,"choices":[{"index":0,"delta":{"content":"9.11.\\n\\n"},"finish_reason":null}]}\n' 2026-01-26 02:26:48.134 | INFO | open_webui.routers.openai:combined_stream:3484 - [STREAM DEBUG] Line #4: b'\n' 2026-01-26 02:26:48.216 | INFO | open_webui.routers.openai:combined_stream:3484 - [STREAM DEBUG] Line #5: b'data: {"id":"chatcmpl-54cba5b4-6db9-4226-9747-c16dd9cb3b44","object":"chat.completion.chunk","created":1769365608216,"choices":[{"index":0,"delta":{"content":"Let me think about this carefully"},"fini' 2026-01-26 02:26:56.066 | INFO | open_webui.routers.openai:combined_stream:3487 - [STREAM DEBUG] Stream ended. Buffered lines: 8, Stream lines: 160 2026-01-26 02:26:56.083 | INFO | open_webui.utils.chat:generate_chat_completion:275 - Model ID: qwen3-30b-a3b, Owned by: openai 2026-01-26 02:26:56.083 | INFO | open_webui.utils.chat:generate_chat_completion:305 - Routing to OpenAI handler - Model ID: qwen3-30b-a3b, owned_by: openai, model keys: ['id', 'name', 'owned_by', 'openai', 'urlIdx', 'connection_type', 'info', 'actions', 'filters', 'tags'] 2026-01-26 02:26:56.086 | INFO | open_webui.routers.openai:get_all_models:2075 - get_all_models() 2026-01-26 02:26:56.087 | INFO | open_webui.routers.openai:generate_chat_completion:3024 - [DEBUG] Thinking disabled (budget=0) 2026-01-26 02:26:56.087 | INFO | open_webui.routers.openai:generate_chat_completion:3058 - [DEBUG] Full payload being sent: {"model": "qwen3-30b-a3b", "messages": [{"role": "user", "content": "### Task:\nSuggest 3-5 relevant follow-up questions or prompts that the user might naturally ask next in this conversation as a **user**, based on the chat history, to help continue or deepen the discussion.\n### Guidelines:\n- Write all follow-up questions from the user\u2019s point of view, directed to the assistant.\n- Make questions concise, clear, and directly related to the discussed topic(s).\n- Only suggest follow-ups that make sense given the chat content and do not repeat what was already covered.\n- If the conversation is very short or not specific, suggest more general (but relevant) follow-ups the user might ask.\n- Use the conversation's primary language; default to English if multilingual.\n- Response must be a JSON array of strings, no extra text or formatting.\n### Output:\nJSON format: { \"follow_ups\": [\"Question 1?\", \"Question 2?\", \"Question 3?\"] }\n### Chat History:\n<chat_history>\nUSER: 9.8\u548c9.11\u54ea\u4e2a\u5927\uff1f\nASSISTANT: 9.8 \u6bd4 9.11 \u5927\u3002\n\n\u6bd4\u8f83\u5c0f\u6570\u65f6\uff0c\u9700\u8981\u4ece\u5de6\u5230\u53f3\u9010\u4f4d\u6bd4\u8f83\uff1a\n- \u4e24\u4e2a\u6570\u7684\u6574\u6570\u90e8\u5206\u90fd\u662f 9\n- \u5728\u5c0f\u6570\u70b9\u540e\u7b2c\u4e00\u4f4d\uff08\u5341\u5206\u4f4d\uff09\uff1a9.8 \u662f 8\uff0c\u800c 9.11 \u662f 1\n- \u56e0\u4e3a 8 > 1\uff0c\u6240\u4ee5 9.8 > 9.11\n\n\u4e5f\u53ef\u4ee5\u8fd9\u6837\u7406\u89e3\uff1a\n- 9.8 = 9.80\n- 9.11 = 9.11\n- 9.80 \u660e\u663e\u5927\u4e8e 9.11\n\n\u6240\u4ee5 9.8 \u66f4\u5927\u3002\n</chat_history>"}], "stream": false, "enable_thinking": true, "include_reasoning": true, "return_reasoning": true, "show_thinking": true}... 2026-01-26 02:26:56.120 | INFO | uvicorn.protocols.http.httptools_impl:send:483 - 192.168.0.1:0 - "POST /api/chat/completed HTTP/1.1" 200 2026-01-26 02:26:56.147 | INFO | uvicorn.protocols.http.httptools_impl:send:483 - 192.168.0.1:0 - "POST /api/v1/chats/273b2d82-2b80-47bb-ba27-f759d6d52931 HTTP/1.1" 200

模型接入建议：

1. 外部连接的模型管理和OpenWebUI的模型管理比较隔离，比如权限组的管理，

平台运营建议：

1. 现阶段可以对模型设置价格计费，但没有对用户可以设置余额管理，期望运营可以增加用户余额管理，看板等。（从开发者角度看，该功能后期可以转商业版）

Actual Behavior

如上

Steps to Reproduce

如上

Logs & Screenshots

如上

Additional Information

No response

[ztx888]

非常感谢佬友的反馈和建议！！！关于思考的BUG已经定位到，会在下个版本修复这个问题！关于权限组，我还在完全重构和优化逻辑整体的逻辑，可能需要后续几个版本才能完美上线。计费方面也会在后续几个版本陆续优化上线。再次感谢建议和反馈！