Update app.py

app.py

@@ -1,70 +1,78 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
 
+# Load Anki-2.5 model
+model_name = "anktechsol/anki-2.5"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype=torch.float32,
+    device_map="auto"
+)
 
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
+def respond(message, history, system_message, max_tokens, temperature, top_p):
     """
-    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
+    Indian Language Q&A powered by Anki-2.5
+    Supports Hindi, Tamil, Bengali, Telugu, and other Indian languages
     """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
-
-    messages = [{"role": "system", "content": system_message}]
-
-    messages.extend(history)
-
-    messages.append({"role": "user", "content": message})
-
-    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
+    # Construct conversation with system message
+    conversation = [{"role": "system", "content": system_message}]
+    
+    # Add history
+    for msg in history:
+        if isinstance(msg, dict):
+            conversation.append(msg)
+        elif len(msg) == 2:
+            conversation.append({"role": "user", "content": msg[0]})
+            conversation.append({"role": "assistant", "content": msg[1]})
+    
+    # Add current message
+    conversation.append({"role": "user", "content": message})
+    
+    # Generate response
+    inputs = tokenizer.apply_chat_template(
+        conversation,
+        tokenize=True,
+        add_generation_prompt=True,
+        return_tensors="pt"
+    ).to(model.device)
+    
+    with torch.no_grad():
+        outputs = model.generate(
+            inputs,
+            max_new_tokens=max_tokens,
+            temperature=temperature,
+            top_p=top_p,
+            do_sample=True,
+            pad_token_id=tokenizer.eos_token_id
+        )
+    
+    response = tokenizer.decode(outputs[0][inputs.shape[-1]:], skip_special_tokens=True)
+    return response
 
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# Create ChatInterface with Indian language focus
 chatbot = gr.ChatInterface(
     respond,
-    type="messages",
+    title="🇮🇳 Indian Language Q&A Assistant - Powered by Anki-2.5",
+    description="Ask questions in Hindi, Tamil, Bengali, Telugu, Urdu, Gujarati, and other Indian languages. This assistant understands Indian cultural context and regional nuances.",
+    examples=[
+        ["भारत की राजधानी क्या है?"],
+        ["இந்தியாவில் எத்தனை மாநிலங்கள் உள்ளன?"],
+        ["ভারতের সবচেয়ে বড় শহর কোনটি?"],
+        ["What are the major festivals celebrated in India?"],
+    ],
     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)",
+        gr.Textbox(
+            value="You are a helpful assistant specializing in Indian languages and culture. You understand Hindi, Tamil, Bengali, Telugu, and other Indian languages. Respond accurately and culturally appropriately.",
+            label="System Message"
         ),
+        gr.Slider(minimum=50, maximum=1024, value=256, step=1, label="Max Tokens"),
+        gr.Slider(minimum=0.1, maximum=1.5, 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"),
     ],
+    theme=gr.themes.Soft(),
 )
 
-with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
-
-
 if __name__ == "__main__":
-    demo.launch()
+    chatbot.launch()