app.py

import streamlit as st
from huggingface_hub import InferenceClient
import os
from typing import Iterator
from PIL import Image
import pytesseract
from PyPDF2 import PdfReader
import base64
from together import Together

API_KEY = os.getenv("TOGETHER_API_KEY")
if not API_KEY:
    raise ValueError("API key is missing! Make sure TOGETHER_API_KEY is set in the Secrets.")


# Initialize the client with Together AI provider
@st.cache_resource
def get_client():
    #return InferenceClient(
    #    provider="together",
    #     api_key=API_KEY
    #)
    return Together(api_key=API_KEY)  # Use Together.ai's official client

def process_file(file) -> str:
    """Process uploaded file and return its content"""
    if file is None:
        return ""

    try:
        # Handle PDF files
        if file.type == "application/pdf":
            text = ""
            pdf_reader = PdfReader(file)
            for page in pdf_reader.pages:
                page_text = page.extract_text()
                if page_text:
                    text += page_text + "\n"
            return text

        # Handle image files
        elif file.type.startswith("image/"):
            return base64.b64encode(file.getvalue()).decode("utf-8")

        # Handle text files
        else:
            return file.getvalue().decode('utf-8')
    except Exception as e:
        return f"Error processing file: {str(e)}"

def generate_response(
    message: str,
    history: list[tuple[str, str]],
    system_message: str,
    max_tokens: int,
    temperature: float,
    top_p: float,
    files=None
) -> Iterator[str]:
    client = get_client()

    has_images = False
    content_blocks = []
    image_content = None  # To store image data
    image_mime_type = None  # To store MIME type

    if files:
        for file in files:
            content = process_file(file)
            if file.type.startswith("image/"):
                has_images = True
                image_content = content  # Already base64 encoded
                image_mime_type = file.type  # Store MIME type
            else:
                content_blocks.append({
                    "type": "text",
                    "text": f"File content:\n{content}"
                })

    # Build messages
    messages = [{"role": "system", "content": system_message}]

    # Add history
    for user_msg, assistant_msg in history:
        messages.append({"role": "user", "content": user_msg})
        messages.append({"role": "assistant", "content": assistant_msg})

    try:
        if has_images:
            # Vision model request
            vision_messages = [{
                "role": "user",
                "content": [
                    {"type": "text", "text": message},
                    {
                        "type": "image_url",
                        "image_url": {
                            "url": f"data:{image_mime_type};base64,{image_content}",
                        },
                    },
                ]
            }]

            stream = client.chat.completions.create(
                model="meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo",
                messages=vision_messages,
                stream=True,
            )

        else:
            # Text-only model request
            current_message = {
                "role": "user",
                "content": [{"type": "text", "text": message}] + content_blocks
            }
            messages.append(current_message)

            stream = client.chat.completions.create(
                model="deepseek-ai/DeepSeek-R1",
                messages=messages,
                max_tokens=max_tokens,
                temperature=temperature,
                top_p=top_p,
                stream=True
            )

        # Stream response
        for chunk in stream:
            if chunk.choices and chunk.choices[0].delta.content:
                yield chunk.choices[0].delta.content

    except Exception as e:
        yield f"Error: {str(e)}"
def main():
    st.set_page_config(page_title="DeepSeek Chat", page_icon="💭", layout="wide")
    
    # Initialize session state for chat history
    if "messages" not in st.session_state:
        st.session_state.messages = []

    st.title("DeepSeek/Llama Vision Chat with File Upload")
    st.markdown("Chat with DeepSeek AI model. You can optionally upload files for the model to analyze.")
    st.markdown("Feel free to upload images too, in this case Llama Vision will be used")

    # Sidebar for parameters
    with st.sidebar:
        st.header("Settings")
        system_message = st.text_area(
            "System Message",
            value="You are a friendly Chatbot.",
            height=100
        )
        max_tokens = st.slider(
            "Max Tokens",
            min_value=1,
            max_value=8192,
            value=8192,
            step=1
        )
        temperature = st.slider(
            "Temperature",
            min_value=0.1,
            max_value=4.0,
            value=0.0,
            step=0.1
        )
        top_p = st.slider(
            "Top-p (nucleus sampling)",
            min_value=0.1,
            max_value=1.0,
            value=0.95,
            step=0.05
        )
        uploaded_file = st.file_uploader(
            "Upload File (optional)",
            type=['txt', 'py', 'md', 'swift', 'java', 'js', 'ts', 'rb', 'go', 
                 'php', 'c', 'cpp', 'h', 'hpp', 'cs', 'html', 'css', 'kt', 'svelte',
                 'pdf', 'png', 'jpg', 'jpeg'],  # Added file types
            accept_multiple_files=True
        )

    # Display chat messages
    for message in st.session_state.messages:
        with st.chat_message(message["role"]):
            st.write(message["content"])

    # Chat input
    if prompt := st.chat_input("What would you like to know?"):
        # Display user message
        st.session_state.messages.append({"role": "user", "content": prompt})
        with st.chat_message("user"):
            st.write(prompt)

        # Generate and display assistant response
        with st.chat_message("assistant"):
            response_placeholder = st.empty()
            full_response = ""
            
            # Get message history for context
            history = [(msg["content"], next_msg["content"]) 
                      for msg, next_msg in zip(st.session_state.messages[::2], st.session_state.messages[1::2])]
            
            # Stream the response
            for response_chunk in generate_response(
                prompt,
                history,
                system_message,
                max_tokens,
                temperature,
                top_p,
                uploaded_file
            ):
                full_response += response_chunk
                print(full_response)
                response_placeholder.markdown(full_response + "▌")
            
            response_placeholder.markdown(full_response)
        
        # Add assistant response to chat history
        st.session_state.messages.append({"role": "assistant", "content": full_response})

if __name__ == "__main__":
    main()