Create app.py

app.py

@@ -0,0 +1,23 @@
+import gradio as gr
+import subprocess
+
+MODEL_PATH = "/app/model/model.gguf"
+
+def query_model(prompt):
+    cmd = [
+        "/llama.cpp/main", "-m", MODEL_PATH,
+        "-t", "4",            # threads
+        "--prompt", prompt,
+        "--n_predict", "128"  # tokens to generate
+    ]
+    result = subprocess.run(cmd, capture_output=True, text=True)
+    return result.stdout
+
+iface = gr.Interface(
+    fn=query_model,
+    inputs="text",
+    outputs="text",
+    title="GGUF Docker Chatbot"
+)
+
+iface.launch(server_name="0.0.0.0", server_port=7860)