ollama init to common utils

VB_interviews_sandbox.py

@@ -7,33 +7,13 @@ app = marimo.App(width="medium")
 @app.cell
 def _():
     import marimo as mo
-    import requests
-    import ollama
-    from ollama import Client
+    from utils import connect_qumo_ollama
 
-    VM_NAME = 'hiperf-gpu'
-    #VM_NAME = 'ollama-vb'
+    # VM_NAME = 'hiperf-gpu'
+    VM_NAME = 'ollama-lite'
 
-    QUMO_OLLAMA_URL = f'http://{VM_NAME}.tail44fa00.ts.net:11434'
-    return Client, QUMO_OLLAMA_URL, VM_NAME, mo, requests
-
-
-@app.cell
-def _(Client, QUMO_OLLAMA_URL, requests):
-    try:
-        requests.get(QUMO_OLLAMA_URL, timeout=5)
-        client = Client(
-            host=QUMO_OLLAMA_URL
-        )
-    except requests.ConnectionError:
-        print(f"Failed to reach {QUMO_OLLAMA_URL}. Check that the VM is running and Tailscale is up")
-    return (client,)
-
-
-@app.cell
-def _():
-    print("Hello")
-    return
+    client = connect_qumo_ollama(VM_NAME)
+    return VM_NAME, client, mo
 
 
 @app.cell(hide_code=True)
@@ -82,7 +62,7 @@ def _(mo):
 
 @app.cell
 def _(client):
-    response_chat = client.chat(model='deepseek-r1:32b', messages=[
+    response_chat = client.chat(model='gemini-3-pro-preview:latest', messages=[
       {
         'role': 'user',
         'content': 'Why is the sky blue?',
@@ -109,7 +89,7 @@ def _(mo):
 
 @app.cell
 def _(client):
-    response_generate = client.generate(model='deepseek-r1:7b', prompt='Why is the sky blue?')
+    response_generate = client.generate(model='deepseek-r1:32b', prompt='Why is the sky blue?')
     return (response_generate,)