Pre-checks
Deployment Method
Source (setup.sh)
Steps to Reproduce
- 安装最新的ollama到本地模型系统中
- 模型池里面配置好相应的ollama参数
- 打开任意数字员工
- 输入“你好“
- 等几分钟后,不管是日志还是对话框都会报错
[LLM call error] [Error] Too many tool call rounds
Expected vs Actual Behavior
期望结果:ollama的本地模型可以被正常调用
实际结果:任意输入都是有问题的
Logs / Screenshots
2026-06-12 13:16:46 | ERROR | b3556b69-1f4 | app.api.websocket:938 - [WS] LLM error after 111.048s: [Error] Too many tool call rounds
Traceback (most recent call last):
File "/clawith/Clawith/backend/.venv/bin/uvicorn", line 8, in
sys.exit(main())
│ │ └
│ └
└ <module 'sys' (built-in)>
File "/clawith/Clawith/backend/.venv/lib64/python3.12/site-packages/click/core.py", line 1485, in call
return self.main(*args, **kwargs)
│ │ │ └ {}
│ │ └ ()
│ └ <function Command.main at 0x7f0e8332e660>
└
File "/clawith/Clawith/backend/.venv/lib64/python3.12/site-packages/click/core.py", line 1406, in main
Pre-checks
Deployment Method
Source (setup.sh)
Steps to Reproduce
[LLM call error] [Error] Too many tool call rounds
Expected vs Actual Behavior
期望结果:ollama的本地模型可以被正常调用
实际结果:任意输入都是有问题的
Logs / Screenshots
2026-06-12 13:16:46 | ERROR | b3556b69-1f4 | app.api.websocket:938 - [WS] LLM error after 111.048s: [Error] Too many tool call rounds
Traceback (most recent call last):
File "/clawith/Clawith/backend/.venv/bin/uvicorn", line 8, in
sys.exit(main())
│ │ └
│ └
└ <module 'sys' (built-in)>
File "/clawith/Clawith/backend/.venv/lib64/python3.12/site-packages/click/core.py", line 1485, in call
return self.main(*args, **kwargs)
│ │ │ └ {}
│ │ └ ()
│ └ <function Command.main at 0x7f0e8332e660>
└
File "/clawith/Clawith/backend/.venv/lib64/python3.12/site-packages/click/core.py", line 1406, in main