不支持本地Ollama部署的模型调用？

[swy755565055]

本地部署了Ollama运行的qwen2.5模型，
llm_config = {
'model': 'qwen2.5:7b',
'model_server': 'http://127.0.0.1:11434/v1/', # base_url，也称为 api_base
'api_key': 'ollama',
}
但是启动每次都报错：NotImplementedError
定位到源码：modelscope_agent\llm_init_.py

def get_chat_model(model: str, model_server: str, **kwargs) -> BaseChatModel:
"""
model: the model name: such as qwen-max, gpt-4 ...
model_server: the source of model, such as dashscope, openai, modelscope ...
**kwargs: more parameters, such as api_key, api_base
"""
model_type = re.split(r'[-/]', model)[0] # parser qwen / gpt / ...
registered_model_id = f'{model_server}{model_type}'
if registered_model_id in LLM_REGISTRY: # specific model from specific source
return LLM_REGISTRY[registered_model_id](model, model_server, **kwargs)
elif model_server in LLM_REGISTRY: # specific source
return LLM_REGISTRY[model_server](model, model_server, **kwargs)
else:
raise NotImplementedError

LLM_REGISTRY里并没有注册本地的model_server

所以目前是还不支持除了DashScope以外的模型调用？