search

run-llama / multi-agent-concierge

An example of multi-agent orchestration with llama-index
MIT License
263 stars 38 forks source link

FunctionCallingAgentWorker is unable to load llama 3.1 via HuggingFaceLLM #2

Closed kumar19a closed 2 days ago

kumar19a commented 2 months ago

Whenever I load a llama 3.1 via HuggingFaceLLM in the FunctionCallingAgentWorker it shows model does not support function calling api. Is the multi agent concierge workload only executed via OpenAI API calls?

MohmedAAK commented 1 month ago

i test it over AzureOpenAI, it works fine.

some llama-index features are only supported with open-ai model so the code may need some of those features.

logan-markewich commented 2 days ago

It should work with any LlM that extends the FunctionCallingLLM class, or has llm.metadata.is_function_calling_model as True