Update app.py

app.py

@@ -20,7 +20,7 @@ raven_pipeline = pipeline(
 @spaces.GPU
 def process_text(input_text: str) -> str:
     prompt = f"User Query: {input_text}<human_end>"
-    result = raven_pipeline(prompt, temperature=0.001, max_new_tokens=2000, return_full_text=False, do_sample=False)[0]["generated_text"].replace("Call:", "").strip()
+    result = raven_pipeline(prompt, temperature=0.001, max_new_tokens=1200, return_full_text=False, do_sample=False)[0]["generated_text"].replace("Call:", "").strip()
 #   torch.cuda.empty_cache()
     return result