Update app.py

app.py

@@ -99,8 +99,8 @@ os.makedirs("models",exist_ok=True)
 
 logging.info("start download")
 hf_hub_download(
-    repo_id="bartowski/google_gemma-3-1b-it-GGUF",
-    filename="google_gemma-3-1b-it-Q5_K_M.gguf",
+    repo_id="bartowski/google_gemma-3-4b-it-GGUF",
+    filename="google_gemma-3-4b-it-Q4_K_M.gguf",
     local_dir="./models",
 )
 
@@ -148,7 +148,7 @@ class LlamaCppModel(Model):
                 n_batch=8,
                 n_ctx=n_ctx,
                 n_threads=2,
-                n_threads_batch=2,verbose=False
+                n_threads_batch=4,verbose=False
             )
             
         elif repo_id and filename:
@@ -255,7 +255,7 @@ class LlamaCppModel(Model):
             
 
 model = LlamaCppModel(
-        model_path = "models/google_gemma-3-1b-it-Q5_K_M.gguf",
+        model_path = "models/google_gemma-3-4b-it-Q4_K_M",
         n_ctx=8192,verbose=False
     )