xerrors
Error: 部分模型对 stream=false 的支持不完善
1️⃣ 描述一下问题
简单描述一下
弃用model_provider_lite后在重写查询时产生错误，用不同的模型会产生不同的错误. 下面第一张图显示的错误好像是百炼密钥api相关的错误，但我的密钥应该没问题，这个不太清楚，麻烦你看一下。
第二张图显示的错误是有些模型只支持流式输出， rewritten_query = model.predict(rewritten_query_prompt).content 这里应该需要加参数stream=true，而且还可能引发其他错误，比如 AttributeError: 'generator' object has no attribute 'content'
2️⃣ 报错日志
请运行以下命令，并提供部分相关日志：
2025-05-07 17:57:46,229 - ERROR - LLM - Retriever error: Error code: 400 - {'error': {'code': 'invalid_parameter_error', 'param': None, 'message': 'current user api does not support http call', 'type': 'invalid_request_error'}, 'id': 'chatcmpl-bdb5ec5d-58c6-9f92-917c-cdadb4369dcc', 'request_id': 'bdb5ec5d-58c6-9f92-917c-cdadb4369dcc'}, Traceback (most recent call last): File "E:\PycharmProjects\llms-local\src\routers\chat_router.py", line 61, in generate_response modified_query, refs = retriever(modified_query, history_manager.messages, meta) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\PycharmProjects\llms-local\src\core\retriever.py", line 184, in call refs = self.retrieval(query, history, meta) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\PycharmProjects\llms-local\src\core\retriever.py", line 24, in retrieval refs["knowledge_base"] = self.query_knowledgebase(query, history, refs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\PycharmProjects\llms-local\src\core\retriever.py", line 107, in query_knowledgebase rw_query = self.rewrite_query(query, history, refs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\PycharmProjects\llms-local\src\core\retriever.py", line 154, in rewrite_query rewritten_query = model.predict(rewritten_query_prompt).content ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\PycharmProjects\llms-local\src\models\chat_model.py", line 140, in predict return self._get_response(messages) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\PycharmProjects\llms-local\src\models\chat_model.py", line 159, in _get_response response = self.client.chat.completions.create( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Anaconda3\envs\llm\Lib\site-packages\openai_utils_utils.py", line 279, in wrapper return func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "E:\Anaconda3\envs\llm\Lib\site-packages\openai\resources\chat\completions\completions.py", line 914, in create return self._post( ^^^^^^^^^^^ File "E:\Anaconda3\envs\llm\Lib\site-packages\openai_base_client.py", line 1242, in post return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Anaconda3\envs\llm\Lib\site-packages\openai_base_client.py", line 919, in request return self._request( ^^^^^^^^^^^^^^ File "E:\Anaconda3\envs\llm\Lib\site-packages\openai_base_client.py", line 1023, in _request raise self._make_status_error_from_response(err.response) from None openai.BadRequestError: Error code: 400 - {'error': {'code': 'invalid_parameter_error', 'param': None, 'message': 'current user api does not support http call', 'type': 'invalid_request_error'}, 'id': 'chatcmpl-bdb5ec5d-58c6-9f92-917c-cdadb4369dcc', 'request_id': 'bdb5ec5d-58c6-9f92-917c-cdadb4369dcc'}
3️⃣ 相关截图
#️⃣ 其他相关信息
✅ 如果问题与模型调用相关，请尝试切换到其他在线模型
