Update app.py

app.py

@@ -36,7 +36,7 @@ tokenizer = AutoTokenizer.from_pretrained("THUDM/LongWriter-glm4-9b",trust_remot
 
 
 @spaces.GPU()
-def stream_chat(message: str, history: list, temperature: float, max_length: int):
+def stream_chat(message: str, history: list, temperature: float, max_new_tokens: int):
     print(f'message is - {message}')
     print(f'history is - {history}')
     conversation = []
@@ -51,13 +51,12 @@ def stream_chat(message: str, history: list, temperature: float, max_length: int
     streamer = TextIteratorStreamer(tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True)
 
     generate_kwargs = dict(
-        max_length=max_length,
+        max_new_tokens=max_new_tokens,
         streamer=streamer,
         do_sample=True,
         top_k=1,
         temperature=temperature,
-        repetition_penalty=1.2,
-        num_beams=1,
+        repetition_penalty=1,
     )
     gen_kwargs = {**input_ids, **generate_kwargs}
 
@@ -88,11 +87,11 @@ with gr.Blocks(css=CSS) as demo:
                 render=False,
             ),
             gr.Slider(
-                minimum=128,
+                minimum=1024,
                 maximum=32768,
                 step=1,
                 value=4096,
-                label="Max Length",
+                label="Max New Tokens",
                 render=False,
             ),
         ],