[BUG]LLM creat bug

[aylitat]

Description

While I use LLM to creat llm,it show me this error:

litellm.exceptions.BadRequestError: litellm.BadRequestError: LLM Provider NOT provided. Pass in the LLM provider you are trying to call. You passed model=moonshot-v1-32k Pass model as E.g. For 'Huggingface' inference endpoints pass incompletion(model='huggingface/starcoder',..)`

code:

llm = get_llm( api_key=BASE_CONFIG.MOONSHOT_API_KEY, base_url=BASE_CONFIG.BASE_URL, model=BASE_CONFIG.MODEL )

But I use the custom-model example: llm = LLM( model="custom-model-name", base_url="https://api.your-provider.com/v1", api_key="your-api-key" ) agent = Agent(llm=llm, ...)

Steps to Reproduce

https://docs.crewai.com/how-to/LLM-Connections/#using-local-models-with-ollama

Expected behavior

pass

Screenshots/Code snippets

llm = LLM( model="custom-model-name", base_url="https://api.your-provider.com/v1", api_key="your-api-key" ) agent = Agent(llm=llm, ...)

Operating System

Windows 10

Python Version

3.10

crewAI Version

lastest

crewAI Tools Version

lastest

Virtual Environment

Venv

Evidence

Possible Solution

None

Additional context

pass

[CStrue]

Have the same issue

crewai = {extras = ["tools"], version = "^0.65.2"}

[gnana70]

Any one found work around?

[CStrue]

For me worked by setting up the LLM for Olama like this:

return LLM( model=os.getenv('OLLAMA_MODEL_LLAMA_31_8B'), temperature=0.0, base_url=os.getenv('OLLAMA_SERVER_REMOTE'), verbose=True )

OLLAMA_SERVER_LOCAL=http://localhost:11434 OLLAMA_MODEL_LLAMA_31_8B=ollama/llama3.1:latest

[aavetis]

based on this issue, looks like you need to prepend the model with openai/ so it believes it is openai api compatible https://github.com/crewAIInc/crewAI/issues/1456