Update app.py

app.py

@@ -14,7 +14,7 @@ DESCRIPTION = """ EXAONE-3.0-7.8B-Instruct Official Demo \
 """
 
 MAX_MAX_NEW_TOKENS = 4096
-DEFAULT_MAX_NEW_TOKENS = 256
+DEFAULT_MAX_NEW_TOKENS = 128
 MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "3840"))
 
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
@@ -30,12 +30,12 @@ model = AutoModelForCausalLM.from_pretrained(
 model.eval()
 
 
-#@spaces.GPU(duration=90)
+@spaces.GPU()
 def generate(
     message: str,
     chat_history: list[tuple[str, str]],
     system_prompt: str,
-    max_new_tokens: int = 256,
+    max_new_tokens: int = 128,
     temperature: float = 0.6,
     top_p: float = 0.9,
     top_k: int = 50,
@@ -63,7 +63,7 @@ def generate(
         gr.Warning(f"Trimmed input from messages as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
     input_ids = input_ids.to(model.device)
 
-    streamer = TextIteratorStreamer(tokenizer, timeout=20.0, skip_prompt=True, skip_special_tokens=True)
+    streamer = TextIteratorStreamer(tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True)
     generate_kwargs = dict(
         {"input_ids": input_ids},
         streamer=streamer,