t41372
运行时“Error processing agent response: list index out of range”
您好，以下是我的配置信息
2025-03-21 11:33:21 | INFO | src.open_llm_vtuber.service_context:init_live2d:156 | Initializing Live2D: shizuku-local 2025-03-21 11:33:21 | INFO | src.open_llm_vtuber.live2d_model:_lookup_model_info:142 | Model Information Loaded. 2025-03-21 11:33:21 | INFO | src.open_llm_vtuber.service_context:init_asr:166 | Initializing ASR: sherpa_onnx_asr 2025-03-21 11:33:21 | INFO | src.open_llm_vtuber.asr.sherpa_onnx_asr:init:81 | Sherpa-Onnx-ASR: Using cpu for inference 2025-03-21 11:33:26 | INFO | src.open_llm_vtuber.service_context:init_tts:178 | Initializing TTS: edge_tts 2025-03-21 11:33:27 | INFO | src.open_llm_vtuber.service_context:init_vad:190 | Initializing VAD: silero_vad 2025-03-21 11:33:48 | INFO | src.open_llm_vtuber.vad.silero:load_vad_model:50 | Loading Silero-VAD model... 2025-03-21 11:33:48 | INFO | src.open_llm_vtuber.service_context:init_agent:202 | Initializing Agent: basic_memory_agent 2025-03-21 11:33:48 | INFO | src.open_llm_vtuber.agent.agent_factory:create_agent:32 | Initializing agent: basic_memory_agent 2025-03-21 11:33:48 | INFO | src.open_llm_vtuber.agent.stateless_llm_factory:create_llm:20 | Initializing LLM: openai_compatible_llm 2025-03-21 11:33:49 | INFO | src.open_llm_vtuber.agent.stateless_llm.openai_compatible_llm:init:51 | Initialized AsyncLLM with the parameters: xxxxx, gpt-4o-mini-2024-07-18 2025-03-21 11:33:49 | INFO | src.open_llm_vtuber.agent.agents.basic_memory_agent:init:66 | BasicMemoryAgent initialized.
但是用户输入时遇到了： 2025-03-21 11:34:11 | INFO | src.open_llm_vtuber.conversations.single_conversation:process_single_conversation:49 | New Conversation Chain 🐼 started! 2025-03-21 11:34:11 | INFO | src.open_llm_vtuber.conversations.single_conversation:process_single_conversation:72 | User input: 你好 2025-03-21 11:34:14 | ERROR | src.open_llm_vtuber.conversations.single_conversation:process_agent_response:154 | Error processing agent response: list index out of range 2025-03-21 11:34:14 | ERROR | src.open_llm_vtuber.conversations.single_conversation:process_single_conversation:112 | Error in conversation chain: list index out of range 请问是什么问题呢？
我也遇到这个问题了，发现是在src\open_llm_vtuber\agent\stateless_llm\openai_compatible_llm.py的92行，async for chunk in stream:中出现chunk.choices为空的情况，可能是因为用了第三方的openai api，返回了空的steam
这个添加那个代码只是屏蔽了空的回复，这个最终是怎么解决？不可以用第三方的openai api吗？
是因为第三方的openai api存在空的回复，屏蔽掉就可以用了