-
-
Notifications
You must be signed in to change notification settings - Fork 1.2k
Open
Labels
area:providerThe bug / feature is about AI Provider, Models, LLM Agent, LLM Agent Runner.The bug / feature is about AI Provider, Models, LLM Agent, LLM Agent Runner.bugSomething isn't workingSomething isn't workingfeature:chatuiThe bug / feature is about astrbot's chatui, webchatThe bug / feature is about astrbot's chatui, webchatplugin-cate:genaiBased on generative AI models such as LLMs or diffusion models.(only startswith [Plugin])Based on generative AI models such as LLMs or diffusion models.(only startswith [Plugin])priority: p0will fix asapwill fix asap
Description
What happened / 发生了什么
使用intern模型时会调用失败 包括不限于
- ModelScope-InternS1-Pro 空回复
- Intern 报错
AstrBot 请求失败。 错误类型: AssertionError 错误信息: 请在平台日志查看和分享错误详情。
使用WebChat无法使用 Aiocq(Onebot11)正常
Reproduce / 如何复现?
Intern免费API https://internlm.intern-ai.org.cn/api
ModelScope API https://www.modelscope.cn/models/Shanghai_AI_Laboratory/Intern-S1-Pro
调用webchat发送信息
AstrBot version, deployment method (e.g., Windows Docker Desktop deployment), provider used, and messaging platform used. / AstrBot 版本、部署方式(如 Windows Docker Desktop 部署)、使用的提供商、使用的消息平台适配器
版本:V14.3
适配器:WebChat
OS
Windows
Logs / 报错日志
Intern官方接口
[20:18:15] [Core] [DBUG] [pipeline.scheduler:88]: pipeline 执行完毕。 [20:18:35] [Core] [DBUG] [webchat.webchat_adapter:207]: WebChatAdapter: [Plain(type=<ComponentType.Plain: 'Plain'>, text='你是谁', convert=True)] [20:18:35] [Core] [INFO] [core.event_bus:59]: [default] [webchat(webchat)] astrbot/astrbot: 你是谁 [20:18:35] [Core] [DBUG] [waking_check.stage:157]: enabled_plugins_name: ['*'] [20:18:35] [Core] [DBUG] [method.star_request:44]: plugin -> session_controller - handle_session_control_agent [20:18:35] [Core] [DBUG] [method.star_request:44]: plugin -> session_controller - handle_empty_mention [20:18:35] [Core] [DBUG] [agent_sub_stages.internal:150]: ready to request llm provider [20:18:35] [Core] [DBUG] [agent_sub_stages.internal:155]: acquired session lock for llm request [20:18:35] [Core] [DBUG] [core.astr_main_agent:423]: Tool set for persona 默认: ['web_search', 'fetch_url', 'web_search_tavily', 'tavily_extract_web_page'] [20:18:35] [Core] [DBUG] [core.astr_main_agent_resources:415]: [知识库] 使用全局配置,知识库数量: 0 [20:18:35] [Core] [DBUG] [pipeline.context_utils:95]: hook(OnLLMRequestEvent) -> astrbot - decorate_llm_req [20:18:35] [Core] [DBUG] [pipeline.context_utils:95]: hook(OnLLMRequestEvent) -> astrbot-web-searcher - edit_web_search_tools [20:18:35] [Core] [INFO] [respond.stage:168]: Prepare to send - astrbot/astrbot: [20:18:35] [Core] [INFO] [respond.stage:184]: 应用流式输出(webchat) [20:18:35] [Core] [DBUG] [runners.base:64]: Agent state transition: AgentState.IDLE -> AgentState.RUNNING [20:18:39] [Core] [ERRO] [v4.14.4] [core.astr_agent_run_util:136]: Traceback (most recent call last):
File "/www/AstrBot/astrbot/core/astr_agent_run_util.py", line 52, in run_agent
async for resp in agent_runner.step():
File "/www/AstrBot/astrbot/core/agent/runners/tool_loop_agent_runner.py", line 183, in step
async for llm_response in self._iter_llm_responses():
File "/www/AstrBot/astrbot/core/agent/runners/tool_loop_agent_runner.py", line 154, in _iter_llm_responses
async for resp in stream: # type: ignore
File "/www/AstrBot/astrbot/core/provider/sources/openai_source.py", line 576, in text_chat_stream
) = await self._handle_api_error(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/www/AstrBot/astrbot/core/provider/sources/openai_source.py", line 465, in _handle_api_error
raise e
File "/www/AstrBot/astrbot/core/provider/sources/openai_source.py", line 564, in text_chat_stream
async for response in self._query_stream(payloads, func_tool):
File "/www/AstrBot/astrbot/core/provider/sources/openai_source.py", line 200, in _query_stream
final_completion = state.get_final_completion()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/www/AstrBot/astrbot-venv/lib/python3.12/site-packages/openai/lib/streaming/chat/_completions.py", line 334, in get_final_completion
chat_completion=self.current_completion_snapshot,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/www/AstrBot/astrbot-venv/lib/python3.12/site-packages/openai/lib/streaming/chat/_completions.py", line 341, in current_completion_snapshot
assert self.__current_completion_snapshot is not None
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AssertionError
ModelScope
[20:19:37] [Core] [DBUG] [webchat.webchat_adapter:207]: WebChatAdapter: [Plain(type=<ComponentType.Plain: 'Plain'>, text='你是谁', convert=True)]
[20:19:37] [Core] [INFO] [core.event_bus:59]: [default] [webchat(webchat)] astrbot/astrbot: 你是谁
[20:19:37] [Core] [DBUG] [waking_check.stage:157]: enabled_plugins_name: ['*']
[20:19:37] [Core] [DBUG] [method.star_request:44]: plugin -> session_controller - handle_session_control_agent
[20:19:37] [Core] [DBUG] [method.star_request:44]: plugin -> session_controller - handle_empty_mention
[20:19:37] [Core] [DBUG] [agent_sub_stages.internal:150]: ready to request llm provider
[20:19:37] [Core] [DBUG] [agent_sub_stages.internal:155]: acquired session lock for llm request
[20:19:37] [Core] [DBUG] [core.astr_main_agent:423]: Tool set for persona 默认: ['web_search', 'fetch_url', 'web_search_tavily', 'tavily_extract_web_page']
[20:19:37] [Core] [DBUG] [core.astr_main_agent_resources:415]: [知识库] 使用全局配置,知识库数量: 0
[20:19:37] [Core] [DBUG] [pipeline.context_utils:95]: hook(OnLLMRequestEvent) -> astrbot - decorate_llm_req
[20:19:37] [Core] [DBUG] [pipeline.context_utils:95]: hook(OnLLMRequestEvent) -> astrbot-web-searcher - edit_web_search_tools
[20:19:37] [Core] [INFO] [respond.stage:168]: Prepare to send - astrbot/astrbot:
[20:19:37] [Core] [INFO] [respond.stage:184]: 应用流式输出(webchat)
[20:19:37] [Core] [DBUG] [runners.base:64]: Agent state transition: AgentState.IDLE -> AgentState.RUNNING
[20:19:40] [Core] [DBUG] [runners.base:64]: Agent state transition: AgentState.RUNNING -> AgentState.DONE
[20:19:40] [Core] [DBUG] [pipeline.context_utils:95]: hook(OnLLMResponseEvent) -> astrbot - record_llm_resp_to_ltm
[20:19:40] [Core] [DBUG] [agent_sub_stages.internal:349]: LLM 响应为空,不保存记录。
[20:19:40] [Core] [INFO] [result_decorate.stage:189]: 流式输出已启用,跳过结果装饰阶段
[20:19:40] [Core] [DBUG] [pipeline.scheduler:88]: pipeline 执行完毕。
Are you willing to submit a PR? / 你愿意提交 PR 吗?
- [√] Yes!
Code of Conduct
- [√] I have read and agree to abide by the project's Code of Conduct。
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
area:providerThe bug / feature is about AI Provider, Models, LLM Agent, LLM Agent Runner.The bug / feature is about AI Provider, Models, LLM Agent, LLM Agent Runner.bugSomething isn't workingSomething isn't workingfeature:chatuiThe bug / feature is about astrbot's chatui, webchatThe bug / feature is about astrbot's chatui, webchatplugin-cate:genaiBased on generative AI models such as LLMs or diffusion models.(only startswith [Plugin])Based on generative AI models such as LLMs or diffusion models.(only startswith [Plugin])priority: p0will fix asapwill fix asap