Update app.py

app.py

@@ -1,119 +1,35 @@
+import os
 import subprocess
-import time
-import requests
-import gradio as gr
-
 from huggingface_hub import hf_hub_download
 
-# =========================
-# CONFIG
-# =========================
-
 MODEL_REPO = "gustajunq/OpenFable-4B-GGUF"
 MODEL_FILE = "openfable-4b-q4_k_m.gguf"
-
 MODEL_ALIAS = "openfable"
-LLAMA_PORT = 8000
-
-# =========================
-# DOWNLOAD MODEL
-# =========================
 
-print("Baixando GGUF...")
+print("Baixando modelo...")
 
 model_path = hf_hub_download(
     repo_id=MODEL_REPO,
     filename=MODEL_FILE,
 )
 
-print("Modelo baixado:", model_path)
-
-# =========================
-# START SERVER
-# =========================
-
-server = subprocess.Popen(
-    [
-        "python",
-        "-m",
-        "llama_cpp.server",
-        "--model",
-        model_path,
-        "--model_alias",
-        MODEL_ALIAS,
-        "--host",
-        "0.0.0.0",
-        "--port",
-        str(LLAMA_PORT),
-    ]
-)
-
-# =========================
-# WAIT FOR READY
-# =========================
-
-def wait_for_server(timeout=300):
-    start = time.time()
-
-    while time.time() - start < timeout:
-        try:
-            r = requests.get(
-                f"http://127.0.0.1:{LLAMA_PORT}/v1/models",
-                timeout=5,
-            )
-
-            if r.status_code == 200:
-                return True
-
-        except Exception:
-            pass
-
-        time.sleep(2)
-
-    return False
-
-ONLINE = wait_for_server()
-
-# =========================
-# UI
-# =========================
-
-def get_status():
-    try:
-        r = requests.get(
-            f"http://127.0.0.1:{LLAMA_PORT}/v1/models",
-            timeout=10,
-        )
-
-        return r.json()
-
-    except Exception as e:
-        return {"error": str(e)}
-
-with gr.Blocks() as demo:
-    gr.Markdown("# OpenFable API")
-
-    if ONLINE:
-        gr.Markdown(
-            f"""
-            ✅ API online
-
-            Modelo: `{MODEL_ALIAS}`
-
-            Endpoint:
-
-            `/v1/chat/completions`
-            """
-        )
-    else:
-        gr.Markdown("❌ Servidor não iniciou.")
-
-    btn = gr.Button("Status")
-    output = gr.JSON()
-
-    btn.click(
-        get_status,
-        outputs=output,
-    )
-
-demo.launch(server_name="0.0.0.0")
+print("Modelo:", model_path)
+
+cmd = [
+    "python",
+    "-m",
+    "llama_cpp.server",
+    "--model",
+    model_path,
+    "--model_alias",
+    MODEL_ALIAS,
+    "--host",
+    "0.0.0.0",
+    "--port",
+    "7860",
+]
+
+print("Iniciando servidor...")
+print(" ".join(cmd))
+
+subprocess.run(cmd)