Update app.py

app.py

@@ -103,7 +103,7 @@ from safetensors.torch import load_file
 
 class ModelInput(BaseModel):
     prompt: str
-    max_new_tokens: int = 50
+    max_new_tokens: int = 2048
 
 app = FastAPI()
 
@@ -144,7 +144,7 @@ except Exception as e:
     print(f"Error during model loading: {e}")
     raise
 
-def generate_response(model, tokenizer, instruction, max_new_tokens=128):
+def generate_response(model, tokenizer, instruction, max_new_tokens=2048):
     """Generate a response from the model based on an instruction."""
     try:
         # Format input for the model