Create app.py

app.py

@@ -0,0 +1,32 @@
+import os
+import subprocess
+from huggingface_hub import hf_hub_download
+
+print("--- START: Inicializacia Space ---")
+
+# 1. Stiahnutie modelu (uvidíš progres v logoch)
+repo_id = "bartowski/Qwen2.5-Coder-7B-Instruct-GGUF"
+filename = "Qwen2.5-Coder-7B-Instruct-Q4_K_M.gguf"
+
+print(f"--- Stahujem model {filename} z {repo_id} ---")
+model_path = hf_hub_download(repo_id=repo_id, filename=filename, local_dir=".")
+print(f"--- Model je pripraveny v: {model_path} ---")
+
+# 2. Spustenie llama-cpp servera
+command = [
+    "python3", "-m", "llama_cpp.server",
+    "--model", filename,
+    "--host", "0.0.0.0",
+    "--port", "7860",
+    "--chat_format", "chatml",
+    "--verbose", "True"
+]
+
+print(f"--- Spustam server: {' '.join(command)} ---")
+
+# Spustíme proces a budeme preposielať logy do konzoly
+process = subprocess.Popen(command, stdout=subprocess.PIPE, stderr=STDOUT, text=True)
+
+# Toto zabezpečí, že každý riadok zo servera uvidíš v logoch okamžite
+for line in process.stdout:
+    print(line, end='')