[Bug]: Perplexity LLM integration incorrectly recognizes chat model as a completion model about llama_index HOT 1 CLOSED

@mathnathan Hello there! I'm here to help you with any bugs, questions, or contributions you have. Let me know how I can assist you while we wait for a human maintainer.

The error occurs because the complete method is not supported for chat models like "llama-3-sonar-large-32k-chat". Instead, you should use the chat method. Here's how you can modify your code:

Replace the Perplexity initialization and query execution with the following:

from llama_index.llms.perplexity import Perplexity
from llama_index.core.llms import ChatMessage

PERPLEXITY_MODEL_NAME = "llama-3-sonar-large-32k-chat"
PERPLEXITY_API_KEY = "pplx-apikey"

perp_llm = Perplexity(
    model=PERPLEXITY_MODEL_NAME,
    api_key=PERPLEXITY_API_KEY,
    temperature=0.5
)

messages_dict = [
    {"role": "system", "content": "Be precise and concise."},
    {"role": "user", "content": "How many rows are there?"}
]
messages = [ChatMessage(**msg) for msg in messages_dict]

response = perp_llm.chat(messages)
print(str(response))

This change ensures that the chat method is used, which is appropriate for chat models ^[1][2].

To continue talking to Dosu, mention @dosu.

from llama_index.