Update app.py

app.py

@@ -1,17 +1,21 @@
 import gradio as gr
-import requests
+import os
+os.system("pip install ollama")
 
-# URL of the Ollama server running on port 8000
-OLLAMA_SERVER_URL = "https://https://broadfield-dev-ollama-server.hf.space/api/predict/"
+from ollama import chat
 
-def chat_with_ollama(prompt):
-    # Send a request to the Ollama server
-    response = requests.post(OLLAMA_SERVER_URL, json={"data": [prompt]})
-    if response.status_code == 200:
-        return response.json()["data"][0]
-    else:
-        return "Error communicating with the Ollama server."
+stream = chat(
+    model='llama3.2',
+    messages=[{'role': 'user', 'content': 'Why is the sky blue?'}],
+    stream=True,
+)
+output=""
+for chunk in stream:
+    print(chunk['message']['content'], end='', flush=True)
+    output+=chunk['message']['content']
+    yield output
 
+    
 # Create a Gradio interface
 iface = gr.Interface(
     fn=chat_with_ollama,