Update app.py

app.py

@@ -48,15 +48,25 @@ huggingface_token = os.getenv("HUGGINGFACE_TOKEN")
 #     local_dir="./models",
 #     token=huggingface_token
 # )
-from huggingface_hub import snapshot_download
-
-snapshot_download(
-    repo_id="unsloth/MiniMax-M2.5-GGUF",
-    repo_type="model",
-    local_dir="./models/minmax",
-    allow_patterns=["UD-IQ1_S/*"],   # 👈 folder inside repo
-    token=huggingface_token      # only if gated/private
+
+
+hf_hub_download(
+    repo_id="unsloth/Qwen3-Coder-Next-GGUF",
+    filename="Qwen3-Coder-Next-Q4_K_S.gguf",
+    local_dir="./models"
 )
+# from huggingface_hub import snapshot_download
+
+# snapshot_download(
+#     repo_id="unsloth/MiniMax-M2.5-GGUF",
+#     repo_type="model",
+#     local_dir="./models/minmax",
+#     allow_patterns=["UD-IQ1_S/*"],   # 👈 folder inside repo
+#     token=huggingface_token      # only if gated/private
+# )
+
+
+
 
 
 # llm = Llama.from_pretrained(
@@ -204,10 +214,10 @@ demo = gr.ChatInterface(
                 "gemma-2-27b-it-Q5_K_M.gguf",
                 # "2b_it_v2.gguf",
                 "GLM-4.7-Flash-UD-Q8_K_XL.gguf",
-                "minmax/UD-IQ1_S/MiniMax-M2.5-UD-IQ1_S-00001-of-00003.gguf",
+                "Qwen3-Coder-Next-Q4_K_S.gguf",
                 "gpt-oss-20b-Q4_K_M.gguf"
             ],
-            value="GLM-4.7-Flash-UD-Q8_K_XL.gguf",
+            value="Qwen3-Coder-Next-Q4_K_S.gguf",
             label="Model",
         ),
         gr.Textbox(