LFM2.5-1.2B: Unlock the context length limit.

app.py

@@ -10,7 +10,6 @@ import gradio as gr
 async def playground(
     message,
     history,
-    num_ctx,
     temperature,
     repeat_penalty,
     top_k,
@@ -41,7 +40,6 @@ async def playground(
         model="hf.co/LiquidAI/LFM2.5-1.2B-Instruct-GGUF:Q4_K_M",
         messages=messages,
         options={
-            "num_ctx": int(num_ctx),
             "temperature": float(temperature),
             "repeat_penalty": float(repeat_penalty),
             "top_k": int(top_k),
@@ -91,15 +89,6 @@ with gr.Blocks(
         )
         gr.Markdown("---")
         gr.Markdown("## Model Parameters")
-        num_ctx = gr.Slider(
-            minimum=512,
-            maximum=1024,
-            value=512,
-            step=128,
-            label="Context Length",
-            info="Maximum context window size (limited to CPU usage)"
-        )
-        gr.Markdown("")
         temperature = gr.Slider(
             minimum=0.1,
             maximum=1.0,
@@ -139,7 +128,6 @@ with gr.Blocks(
     gr.ChatInterface(
         fn=playground,
         additional_inputs=[
-            num_ctx,
             temperature,
             repeat_penalty,
             top_k,