extend max tokens

app.py

@@ -279,7 +279,7 @@ with gr.Blocks(title="LLM Inference with ZeroGPU") as demo:
             search_chk = gr.Checkbox(label="Enable Web Search", value=True)
             sys_prompt = gr.Textbox(label="System Prompt", lines=3, value=update_default_prompt(search_chk.value))
             gr.Markdown("### Generation Parameters")
-            max_tok = gr.Slider(64, 1024, value=512, step=32, label="Max Tokens")
+            max_tok = gr.Slider(64, 16384, value=2048, step=32, label="Max Tokens")
             temp = gr.Slider(0.1, 2.0, value=0.7, step=0.1, label="Temperature")
             k = gr.Slider(1, 100, value=40, step=1, label="Top-K")
             p = gr.Slider(0.1, 1.0, value=0.9, step=0.05, label="Top-P")