Update app.py

app.py

@@ -1,69 +1,79 @@
+import os
 import gradio as gr
-from huggingface_hub import InferenceClient
+from llama_cpp import Llama
+from huggingface_hub import hf_hub_download
 
+# Enable the high-speed Rust-based downloader from your script
+os.environ["HF_HUB_ENABLE_HF_TRANSFER"] = "1"
 
-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")
+# Your Model List from the original script
+MODELS = {
+    "Llama 3.2 1B (Text)": {"repo": "unsloth/Llama-3.2-1B-Instruct-GGUF", "file": "Llama-3.2-1B-Instruct-Q4_K_M.gguf"},
+    "Llama 3.2 3B": {"repo": "bartowski/Llama-3.2-3B-Instruct-GGUF", "file": "Llama-3.2-3B-Instruct-Q4_K_M.gguf"},
+    "Qwen 2.5 1.5B (Code)": {"repo": "Qwen/Qwen2.5-1.5B-Instruct-GGUF", "file": "qwen2.5-1.5b-instruct-q4_k_m.gguf"},
+    "Phi-3.5 Mini (Microsoft)": {"repo": "bartowski/Phi-3.5-mini-instruct-GGUF", "file": "Phi-3.5-mini-instruct-Q4_K_M.gguf"},
+    "DeepSeek R1 1.5B": {"repo": "bartowski/DeepSeek-R1-Distill-Qwen-1.5B-GGUF", "file": "DeepSeek-R1-Distill-Qwen-1.5B-Q4_K_M.gguf"},
+}
 
-    messages = [{"role": "system", "content": system_message}]
+# Global LLM instance
+llm_instance = None
 
-    messages.extend(history)
+def load_and_chat(message, history, hf_token, model_choice):
+    global llm_instance
+    
+    if not hf_token or len(hf_token) < 10:
+        yield "⚠️ Please enter a valid Hugging Face 'Read' Token in the sidebar first."
+        return
 
-    messages.append({"role": "user", "content": message})
+    try:
+        # Download and Initialize only if not already loaded
+        if llm_instance is None:
+            yield f"⏳ Authenticating and downloading {model_choice}..."
+            path = hf_hub_download(
+                repo_id=MODELS[model_choice]["repo"], 
+                filename=MODELS[model_choice]["file"],
+                token=hf_token # Uses the user's specific token
+            )
+            llm_instance = Llama(model_path=path, n_ctx=2048, n_threads=4, verbose=False)
+        
+        # Generation Logic
+        prompt = f"<|user|>\n{message}\n<|assistant|>\n"
+        response = ""
+        for chunk in llm_instance(prompt, max_tokens=512, stream=True):
+            token = chunk["choices"][0]["text"]
+            response += token
+            yield response
+            
+    except Exception as e:
+        yield f"❌ Error: {str(e)}. Check if your token has 'Read' permissions."
 
-    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
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-chatbot = gr.ChatInterface(
-    respond,
-    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()
+# UI Layout
+with gr.Blocks(theme=gr.themes.Default(primary_hue="blue")) as demo:
+    gr.Markdown("# 🚀 BEN AI - Secure Local Intelligence")
+    
+    with gr.Row():
+        with gr.Column(scale=1):
+            gr.Markdown("### 🔑 Authentication")
+            user_hf_token = gr.Textbox(
+                label="Your HF 'Read' Token", 
+                placeholder="paste hf_...", 
+                type="password"
+            )
+            model_select = gr.Dropdown(
+                choices=list(MODELS.keys()), 
+                value="Llama 3.2 1B (Text)", 
+                label="Target Model"
+            )
+            gr.Markdown("---")
+            gr.Markdown("💡 **Note:** Your token is used only for this session to download the model weights.")
 
+        with gr.Column(scale=4):
+            gr.ChatInterface(
+                fn=load_and_chat,
+                additional_inputs=[user_hf_token, model_select],
+                title="Chat Studio",
+                description="Model weights will download once you send your first message with a valid token."
+            )
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch(server_name="0.0.0.0", server_port=7860)