Update app.py

app.py

@@ -22,7 +22,7 @@ MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
 HF_TOKEN = os.getenv("HF_TOKEN")
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
-model_name = "ehristoforu/cq3bitfp32"
+model_name = "datalama/EXAONE-3.5-7.8B-Instruct-Llamafied"
 
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
@@ -33,9 +33,9 @@ tokenizer = AutoTokenizer.from_pretrained(model_name)
 #peft_model = AutoPeftModelForCausalLM.from_pretrained("ehristoforu/CoolQwen2.5-3b-it")
 #merged_model = peft_model.merge_and_unload()
 #merged_model.save_pretrained("./coolqwen")
-model.save_pretrained("./coolqwen")
-tokenizer.save_pretrained("./coolqwen")
-
+#model.save_pretrained("./coolqwen")
+#tokenizer.save_pretrained("./coolqwen")
+'''
 from huggingface_hub import HfApi
 
 api = HfApi()
@@ -48,8 +48,9 @@ api.upload_folder(
     repo_type="model",
     token=HF_TOKEN,
 )
+'''
 
-@spaces.GPU(duration=240)
+@spaces.GPU(duration=60)
 def generate(
     message: str,
     chat_history: list[tuple[str, str]],
@@ -108,7 +109,7 @@ chat_interface = gr.ChatInterface(
         ),
         gr.Slider(
             label="Temperature",
-            minimum=0.1,
+            minimum=0.0,
             maximum=4.0,
             step=0.1,
             value=0.6,