Update app.py

app.py

@@ -1,70 +1,103 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from llama_cpp import Llama
+import os
 
+# =========================
+# CONFIG
+# =========================
+MODEL_PATH = "model/dogeai-v2-4b-reasoning.Q4.gguf"
 
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
-    """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-    """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
+N_CTX = 4096
+N_THREADS = max(1, os.cpu_count() - 1)
 
-    messages = [{"role": "system", "content": system_message}]
+llm = None  # lazy loading
 
-    messages.extend(history)
 
-    messages.append({"role": "user", "content": message})
+# =========================
+# LOAD MODEL (LAZY)
+# =========================
+def load_model():
+    global llm
+    if llm is None:
+        llm = Llama(
+            model_path=MODEL_PATH,
+            n_ctx=N_CTX,
+            n_threads=N_THREADS,
+            verbose=False
+        )
+    return llm
 
-    response = ""
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
+# =========================
+# REASONING PROMPT
+# =========================
+def build_prompt(user_input):
+    return f"""
+You are DogeAI-v2.0-4B-Reasoning.
 
-        response += token
-        yield response
+Follow this structure STRICTLY:
+1. Analyze the problem step by step.
+2. Decompose assumptions.
+3. Reason explicitly.
+4. Then provide a concise final answer.
 
+Do NOT skip reasoning.
 
+User question:
+{user_input}
+
+Response:
 """
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-chatbot = gr.ChatInterface(
-    respond,
-    type="messages",
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
-
-with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
-
-
-if __name__ == "__main__":
-    demo.launch()
+
+
+# =========================
+# CHAT FUNCTION
+# =========================
+def chat(user_input):
+    model = load_model()
+
+    prompt = build_prompt(user_input)
+
+    output = model(
+        prompt,
+        max_tokens=512,
+        temperature=0.7,
+        top_p=0.9,
+        stop=["User:"]
+    )
+
+    text = output["choices"][0]["text"].strip()
+    return text
+
+
+# =========================
+# GRADIO UI
+# =========================
+with gr.Blocks(title="DogeAI-v2.0-4B-Reasoning") as demo:
+    gr.Markdown(
+        "# 🐕 DogeAI-v2.0-4B-Reasoning\n"
+        "Reasoning-first language model running fully on CPU.\n\n"
+        "**Quantized 4-bit · Explicit thinking · HF Space demo**"
+    )
+
+    input_box = gr.Textbox(
+        label="Your question",
+        placeholder="Ask something that requires reasoning...",
+        lines=4
+    )
+
+    output_box = gr.Textbox(
+        label="DogeAI Response",
+        lines=12
+    )
+
+    run_btn = gr.Button("Think 🧠")
+
+    run_btn.click(
+        fn=chat,
+        inputs=input_box,
+        outputs=output_box
+    )
+
+
+demo.launch()