Invalid tool_choice type: 'object'. Supported string values: none, auto, required"

[snow-ghost]

Agent: sgr_vampi_code
Session ID: ...c12ebb16ac3d

❌ Agent execution error: Error code: 400 - {'error': "Invalid tool_choice type: 'object'. Supported string values: none, auto, required"}
Traceback (most recent call last):
  File "/opt/projects/github.com/vamplabAI/sgr-vampi-code/sgr_deep_research/core/agents/base_agent.py", line 183, in execute
    reasoning = await self._reasoning_phase()
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/projects/github.com/vamplabAI/sgr-vampi-code/sgr_deep_research/core/agents/sgr_vampi_code_agent.py", line 179, in _reasoning_phase
    async with self.openai_client.chat.completions.stream(**request_kwargs) as stream:
  File "/opt/projects/github.com/vamplabAI/sgr-vampi-code/.venv/lib/python3.12/site-packages/openai/lib/streaming/chat/_completions.py", line 273, in __aenter__
    raw_stream = await self.__api_request
                 ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/projects/github.com/vamplabAI/sgr-vampi-code/.venv/lib/python3.12/site-packages/openai/resources/chat/completions/completions.py", line 2672, in create

Use example from readme, run uv sync, fix config.yaml to:

openai:
  api_key: "your-openai-api-key-here"  # Required: Your OpenAI API key
  base_url: "http://localhost:1234/v1/"                         # Optional: Alt>
  model: "qwen/qwen3-4b-2507"                 # Model to use

and run chat mode, everything typing and press Retun