Create app.py

app.py

@@ -0,0 +1,45 @@
+import os
+import gradio as gr
+from huggingface_hub import hf_hub_download
+from llama_cpp import Llama
+
+# Model config
+REPO_ID = "darkc0de/XORTRON-XPRT3-FAST"
+FILENAME = "XORTRON-XPRT3-Q1_S.gguf"  # Quantized model size ~899MB
+
+print("Downloading model from HF Hub...")
+model_path = hf_hub_download(repo_id=REPO_ID, filename=FILENAME)
+
+print("Loading GGUF model on CPU...")
+llm = Llama(
+    model_path=model_path,
+    n_threads=2,  # Free CPU tier me 2 threads optimized rehte hain
+    n_ctx=2048,
+    verbose=False
+)
+
+def generate_response(message, history, max_tokens, temperature):
+    prompt = f"User: {message}\nAssistant:"
+    
+    output = llm(
+        prompt,
+        max_tokens=int(max_tokens),
+        temperature=float(temperature),
+        stop=["User:", "\n\n"]
+    )
+    
+    return output["choices"][0]["text"].strip()
+
+demo = gr.ChatInterface(
+    fn=generate_response,
+    additional_inputs=[
+        gr.Slider(minimum=16, maximum=1024, value=256, step=16, label="Max Tokens"),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.7, step=0.1, label="Temperature"),
+    ],
+    title="XORTRON-XPRT3 (Docker CPU)",
+    description="Running GGUF Quantized Model via Docker on HF Free CPU."
+)
+
+if __name__ == "__main__":
+    # HF Docker environment ke liye server_name="0.0.0.0" aur port=7860 zaroori hai
+    demo.launch(server_name="0.0.0.0", server_port=7860)