Does anyone know how to resolve this issue or if it’s even possible to work around it? I was working with a notebook where I used llm=OpenAI(model=gpt-4), then I tried with llama3-70b-8192 from Groq, and encountered this problem.

Here’s the code snippet:

from llama_index.llms.groq import Groq
from llama_index.core import Settings

llm = Groq(api_key=GROQ_API_KEY, model="llama3-70b-8192")

from llama_parse import LlamaParse

pdf_file_name = '/kaggle/input/pro3-u2-teo-programacionladocliente/TEO_U2_ProgramacionDelLadoDelCliente_clean.pdf'

documents = LlamaParse(api_key=LLAMA_CLOUD_API_KEY, result_type="markdown").load_data(pdf_file_name)

from llama_index.core.node_parser import MarkdownElementNodeParser

node_parser = MarkdownElementNodeParser(llm=Groq(api_key=GROQ_API_KEY, model="llama3-70b-8192"), num_workers=8)

nodes = node_parser.get_nodes_from_documents(documents)

And this is the error message:

ValueError: The model name 'llama3-70b-8192' does not support function calling API.

Thanks in advanced!

I tried to use Groq with MarkdownElementNodeParser and encountered this error. I also tried Gemini, which worked.