replacing deepseek-r1 with llama-3.1

app.py

@@ -3,6 +3,7 @@ import gradio as gr
 from sentence_transformers import SentenceTransformer
 import torch
 from huggingface_hub import InferenceClient
+from groq import Groq
 
 financialText = ""
 file_names = [
@@ -38,7 +39,7 @@ def getTopChunks(query, chunkEmbeddings, textChunks):
 cleanedChunks = preprocessText(financialText)
 chunkEmbeddings = createEmbeddings(cleanedChunks)
 
-client = InferenceClient("deepseek-ai/DeepSeek-R1", token = os.environ.get("SF_TOKEN"))
+client = Groq(api_key = os.environ.get("SF_TOKEN"))
 
 def respond(message, history):
     messages = [{"role": "system",
@@ -63,7 +64,7 @@ def respond(message, history):
     response = ""
 
     responseStream = client.chat_completion(
-        messages, stream = True, max_tokens = 3000, temperature = 0.4
+        model = "llama-3.1-8b-instant", messages, stream = True, max_tokens = 600, temperature = 0.4
     )
     
     for segment in responseStream: