app.py

import aiofiles
import asyncio
import base64
import fitz
import glob
import logging
import os
import pandas as pd
import pytz
import random
import re
import requests
import shutil
import streamlit as st
import time
import torch
import zipfile

from dataclasses import dataclass
from datetime import datetime
from diffusers import StableDiffusionPipeline
from io import BytesIO
from openai import OpenAI
from PIL import Image
from transformers import AutoModelForCausalLM, AutoTokenizer, AutoModel
from typing import Optional

# 🤖 OpenAI wizardry: Summon your API magic!
client = OpenAI(
    api_key=os.getenv('OPENAI_API_KEY'),
    organization=os.getenv('OPENAI_ORG_ID')
)

# 📜 Logging activated: Capturing chaos and calm!
logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")
logger = logging.getLogger(__name__)
log_records = []
class LogCaptureHandler(logging.Handler):
    def emit(self, record):
        log_records.append(record)
logger.addHandler(LogCaptureHandler())

# 🎨 Streamlit styling: Designing a cosmic interface!
st.set_page_config(
    page_title="AI Vision & SFT Titans 🚀",
    page_icon="🤖",
    layout="wide",
    initial_sidebar_state="expanded",
    menu_items={
        'Get Help': 'https://huggingface.co/awacke1',
        'Report a Bug': 'https://huggingface.co/spaces/awacke1',
        'About': "AI Vision & SFT Titans: PDFs, OCR, Image Gen, Line Drawings, Custom Diffusion, and SFT on CPU! 🌌"
    }
)

st.session_state.setdefault('history', [])  # 🌱 History: starting fresh if empty!
st.session_state.setdefault('builder', None)  # 🛠️ Builder: set up the builder if it's missing!
st.session_state.setdefault('model_loaded', False)  # 🚦 Model Loaded: mark as not loaded by default!
st.session_state.setdefault('processing', {})  # ⏳ Processing: initialize processing state as an empty dict!
st.session_state.setdefault('asset_checkboxes', {})  # ✅ Asset Checkboxes: default to an empty dictionary!
st.session_state.setdefault('downloaded_pdfs', {})  # 📄 Downloaded PDFs: start with no PDFs downloaded!
st.session_state.setdefault('unique_counter', 0)  # 🔢 Unique Counter: initialize the counter to zero!
st.session_state.setdefault('selected_model_type', "Causal LM")  # 🧠 Selected Model Type: default to "Causal LM"!
st.session_state.setdefault('selected_model', "None")  # 🤖 Selected Model: set to "None" if not already set!
st.session_state.setdefault('cam0_file', None)  # 📸 Cam0 File: no file loaded by default!
st.session_state.setdefault('cam1_file', None)  # 📸 Cam1 File: no file loaded by default!

# Create a single placeholder for the asset gallery in the sidebar.
if 'asset_gallery_container' not in st.session_state:
    st.session_state['asset_gallery_container'] = st.sidebar.empty()

@dataclass  # 🎨 ModelConfig: A blueprint for model configurations!
class ModelConfig:
    name: str
    base_model: str
    size: str
    domain: Optional[str] = None
    model_type: str = "causal_lm"
    @property
    def model_path(self): return f"models/{self.name}"  # 🚀 Model Path: Home base for brilliance!

@dataclass  # 🎨 DiffusionConfig: Where diffusion magic takes shape!
class DiffusionConfig:
    name: str
    base_model: str
    size: str
    domain: Optional[str] = None
    @property
    def model_path(self): return f"diffusion_models/{self.name}"  # 🚀 Diffusion Path: Let the diffusion begin!

class ModelBuilder:  # 🔧 ModelBuilder: Crafting AI wonders with wit!
    def __init__(self):  # 🚀 Initialize: Setting up the AI factory!
        self.config = None  # No config yet—waiting for genius!
        self.model = None  # Model not built until the magic happens!
        self.tokenizer = None  # Tokenizer: Ready to speak in AI!
        self.jokes = [  # 🤣 Jokes to keep the circuits laughing!
            "Why did the AI go to therapy? Too many layers to unpack! 😂",
            "Training complete! Time for a binary coffee break. ☕",
            "I told my neural network a joke; it couldn't stop dropping bits! 🤖",
            "I asked the AI for a pun, and it said, 'I'm punning on parallel processing!' 😄",
            "Debugging my code is like a stand-up routine—always a series of exceptions! 😆"
        ]
    def load_model(self, model_path: str, config: Optional[ModelConfig] = None):  # 🔄 load_model: Booting up genius!
        with st.spinner(f"Loading {model_path}... ⏳"):  # ⏳ Spinner: Genius loading...
            self.model = AutoModelForCausalLM.from_pretrained(model_path)
            self.tokenizer = AutoTokenizer.from_pretrained(model_path)
            if self.tokenizer.pad_token is None: 
                self.tokenizer.pad_token = self.tokenizer.eos_token  # 🔧 Fix pad token if missing!
            if config: 
                self.config = config  # 🛠️ Config loaded—setting the stage!
            self.model.to("cuda" if torch.cuda.is_available() else "cpu")  # 💻 Deploying the model to its device!
        st.success(f"Model loaded! 🎉 {random.choice(self.jokes)}")  # 🎉 Success: Model is now in orbit!
        return self
    def save_model(self, path: str):  # 💾 save_model: Securing your masterpiece!
        with st.spinner("Saving model... 💾"):  # ⏳ Spinner: Saving brilliance...
            os.makedirs(os.path.dirname(path), exist_ok=True)
            self.model.save_pretrained(path)
            self.tokenizer.save_pretrained(path)  # 📂 Directory magic: Creating and saving!
        st.success(f"Model saved at {path}! ✅")  # ✅ Success: Your model is safely stored!

class DiffusionBuilder:
    def __init__(self):
        self.config = None
        self.pipeline = None
    def load_model(self, model_path: str, config: Optional[DiffusionConfig] = None):
        with st.spinner(f"Loading diffusion model {model_path}... ⏳"):
            self.pipeline = StableDiffusionPipeline.from_pretrained(model_path, torch_dtype=torch.float32).to("cpu")
            if config:
                self.config = config
        st.success("Diffusion model loaded! 🎨")
        return self
    def save_model(self, path: str):
        with st.spinner("Saving diffusion model... 💾"):
            os.makedirs(os.path.dirname(path), exist_ok=True)
            self.pipeline.save_pretrained(path)
        st.success(f"Diffusion model saved at {path}! ✅")
    def generate(self, prompt: str):
        return self.pipeline(prompt, num_inference_steps=20).images[0]

def generate_filename(sequence, ext="png"): 
    return f"{sequence}_{time.strftime('%d%m%Y%H%M%S')}.{ext}"  # ⏳ Generate filename with timestamp magic!

def pdf_url_to_filename(url):
    return re.sub(r'[<>:"/\\|?*]', '_', url) + ".pdf"  # 📄 Convert URL to a safe PDF filename – no hackers allowed!

def get_download_link(file_path, mime_type="application/pdf", label="Download"):
    return f'<a href="data:{mime_type};base64,{base64.b64encode(open(file_path, "rb").read()).decode()}" download="{os.path.basename(file_path)}">{label}</a>'  # 🔗 Create a download link – click it like it's hot!

def zip_directory(directory_path, zip_path): 
    with zipfile.ZipFile(zip_path, 'w', zipfile.ZIP_DEFLATED) as zipf:
        [zipf.write(os.path.join(root, file), os.path.relpath(os.path.join(root, file), os.path.dirname(directory_path)))
         for root, _, files in os.walk(directory_path) for file in files]  # 🎁 Zip directory: Packing files faster than Santa on Christmas Eve!

def get_model_files(model_type="causal_lm"):
    return [d for d in glob.glob("models/*" if model_type == "causal_lm" else "diffusion_models/*") if os.path.isdir(d)] or ["None"]  # 📂 Get model files: Hunting directories like a pro!

def get_gallery_files(file_types=["png", "pdf"]):
    return sorted(list({f for ext in file_types for f in glob.glob(f"*.{ext}")}))  # 🖼️ Get gallery files: Finding art in a digital haystack!

def get_pdf_files():
    return sorted(glob.glob("*.pdf"))  # 📄 Get PDF files: Sorted and served – no paper cuts here!

# 📥 Download PDF: Delivering docs faster than a caffeinated courier!
def download_pdf(url, output_path):
    try: 
        response = requests.get(url, stream=True, timeout=10)
        if response.status_code == 200:
            with open(output_path, "wb") as f:
                for chunk in response.iter_content(chunk_size=8192):
                    f.write(chunk)
            ret = True
        else:
            ret = False
    except requests.RequestException as e: 
        logger.error(f"Failed to download {url}: {e}")
        ret = False
    return ret  

# 📚 Async PDF Snapshot: Snap your PDF pages without blocking—juggle pages like a ninja! 🥷
async def process_pdf_snapshot(pdf_path, mode="single"):  
    start_time = time.time()
    status = st.empty()
    status.text(f"Processing PDF Snapshot ({mode})... (0s)")
    try:
        doc = fitz.open(pdf_path)
        output_files = []
        if mode == "single":
            page = doc[0]
            pix = page.get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
            output_file = generate_filename("single", "png")
            pix.save(output_file)
            output_files.append(output_file)
        elif mode == "twopage":
            for i in range(min(2, len(doc))):
                page = doc[i]
                pix = page.get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
                output_file = generate_filename(f"twopage_{i}", "png")
                pix.save(output_file)
                output_files.append(output_file)
        elif mode == "allpages":
            for i in range(len(doc)):
                page = doc[i]
                pix = page.get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
                output_file = generate_filename(f"page_{i}", "png")
                pix.save(output_file)
                output_files.append(output_file)
        doc.close()
        elapsed = int(time.time() - start_time)
        status.text(f"PDF Snapshot ({mode}) completed in {elapsed}s!")
        update_gallery()
        return output_files
    except Exception as e:
        status.error(f"Failed to process PDF: {str(e)}")
        return []

# 😎 Async OCR: Convert images to text while your app keeps on groovin'—no blocking, just rocking! 🎸
async def process_ocr(image, output_file):  
    start_time = time.time()
    status = st.empty()
    status.text("Processing GOT-OCR2_0... (0s)")
    tokenizer = AutoTokenizer.from_pretrained("ucaslcl/GOT-OCR2_0", trust_remote_code=True)
    model = AutoModel.from_pretrained("ucaslcl/GOT-OCR2_0", trust_remote_code=True, torch_dtype=torch.float32).to("cpu").eval()
    temp_file = f"temp_{int(time.time())}.png"
    image.save(temp_file)
    result = model.chat(tokenizer, temp_file, ocr_type='ocr')
    os.remove(temp_file)
    elapsed = int(time.time() - start_time)
    status.text(f"GOT-OCR2_0 completed in {elapsed}s!")
    async with aiofiles.open(output_file, "w") as f:
        await f.write(result)
    update_gallery()
    return result

# 🧞 Async Image Gen: Your image genie—wishing up pictures while the event loop keeps the party going! 🎉
async def process_image_gen(prompt, output_file):  
    start_time = time.time()
    status = st.empty()
    status.text("Processing Image Gen... (0s)")
    pipeline = st.session_state['builder'].pipeline if st.session_state.get('builder') and isinstance(st.session_state['builder'], DiffusionBuilder) and st.session_state['builder'].pipeline else StableDiffusionPipeline.from_pretrained("OFA-Sys/small-stable-diffusion-v0", torch_dtype=torch.float32).to("cpu")
    gen_image = pipeline(prompt, num_inference_steps=20).images[0]
    elapsed = int(time.time() - start_time)
    status.text(f"Image Gen completed in {elapsed}s!")
    gen_image.save(output_file)
    update_gallery()
    return gen_image

# 🖼️ GPT-Image Interpreter: Turning pixels into prose!
def process_image_with_prompt(image, prompt, model="gpt-4o-mini", detail="auto"): 
    buffered = BytesIO()
    image.save(buffered, format="PNG")  # 💾 Save the image in-memory as PNG—no hard drives harmed!
    img_str = base64.b64encode(buffered.getvalue()).decode("utf-8")  # 🔐 Encode image data in Base64 for secure, inline transmission!
    messages = [{"role": "user", "content": [
                    {"type": "text", "text": prompt},
                    {"type": "image_url", "image_url": {"url": f"data:image/png;base64,{img_str}", "detail": detail}}
                ]}]
    try:
        response = client.chat.completions.create(model=model, messages=messages, max_tokens=300)
        return response.choices[0].message.content
    except Exception as e:
        return f"Error processing image with GPT: {str(e)}"

# 📝 GPT-Text Alchemist: Merging your prompt and text into digital gold!
def process_text_with_prompt(text, prompt, model="gpt-4o-mini"):  
    messages = [{"role": "user", "content": f"{prompt}\n\n{text}"}]
    try: 
        response = client.chat.completions.create(model=model, messages=messages, max_tokens=300)
        return response.choices[0].message.content
    except Exception as e:
        return f"Error processing text with GPT: {str(e)}"

st.sidebar.subheader("Gallery Settings")  # 🎨 Sidebar Gallery: Customize your creative space!
st.session_state.setdefault('gallery_size', 2)  # 🔧 Setting default gallery size to 2 if it's missing!
st.session_state['gallery_size'] = st.sidebar.slider("Gallery Size", 1, 10, st.session_state['gallery_size'], key="gallery_size_slider")  # 🎚️ Slide to adjust your gallery size and bring balance to your art!

# 📸 Gallery Updater: Making your assets dazzle and disappear faster than a magician's rabbit! 🐇✨
def update_gallery():
    container = st.session_state['asset_gallery_container']
    container.empty()  # Clear previous gallery content
    all_files = get_gallery_files()  # 🔍 Grab all gallery files like a digital treasure hunt!
    if all_files:
        with container:
            st.sidebar.subheader("Asset Gallery 📸📖")
            cols = st.sidebar.columns(2)
            for idx, file in enumerate(all_files[:st.session_state['gallery_size']]):
                with cols[idx % 2]:
                    st.session_state['unique_counter'] += 1
                    unique_id = st.session_state['unique_counter']
                    if file.endswith('.png'):
                        st.image(Image.open(file), caption=os.path.basename(file), use_container_width=True)
                    else:
                        doc = fitz.open(file)
                        pix = doc[0].get_pixmap(matrix=fitz.Matrix(0.5, 0.5))
                        img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
                        st.image(img, caption=os.path.basename(file), use_container_width=True)
                        doc.close()
                    checkbox_key = f"asset_{file}_{unique_id}"
                    st.session_state['asset_checkboxes'][file] = st.checkbox("Use for SFT/Input", value=st.session_state['asset_checkboxes'].get(file, False), key=checkbox_key)
                    mime_type = "image/png" if file.endswith('.png') else "application/pdf"
                    st.markdown(get_download_link(file, mime_type, "Snag It! 📥"), unsafe_allow_html=True)
                    if st.button("Zap It! 🗑️", key=f"delete_{file}_{unique_id}"):
                        os.remove(file)
                        st.session_state['asset_checkboxes'].pop(file, None)
                        st.sidebar.success(f"Asset {os.path.basename(file)} vaporized! 💨")
                        st.rerun()

st.sidebar.subheader("Action Logs 📜")  # 📝 Action Logs: Where our system whispers its secrets!
with st.sidebar:
    [st.write(f"{record.asctime} - {record.levelname} - {record.message}") for record in log_records]

st.sidebar.subheader("History 📜")  # 🕰️ History: A walk down memory lane, one log at a time!
with st.sidebar:
    [st.write(entry) for entry in st.session_state['history']]

tabs = st.tabs(["Camera Snap 📷", "Download PDFs 📥", "Test OCR 🔍", "Build Titan 🌱", "Test Image Gen 🎨", "PDF Process 📄", "Image Process 🖼️", "MD Gallery 📚"])
(tab_camera, tab_download, tab_ocr, tab_build, tab_imggen, tab_pdf_process, tab_image_process, tab_md_gallery) = tabs

with tab_camera:
    st.header("Camera Snap 📷")  # 🎥 Header: Let’s capture those Kodak moments!
    st.subheader("Single Capture")  # 📸 Subheader: One snap at a time, no double exposure!
    cols = st.columns(2)
    
    with cols[0]:
        cam0_img = st.camera_input("Take a picture - Cam 0", key="cam0")
        if cam0_img:
            filename = generate_filename("cam0")
            if st.session_state['cam0_file'] and os.path.exists(st.session_state['cam0_file']):
                os.remove(st.session_state['cam0_file'])
            with open(filename, "wb") as f:
                f.write(cam0_img.getvalue())
            st.session_state['cam0_file'] = filename
            entry = f"Snapshot from Cam 0: {filename}"
            if entry not in st.session_state['history']:
                st.session_state['history'] = [e for e in st.session_state['history'] if not e.startswith("Snapshot from Cam 0:")] + [entry]
            st.image(Image.open(filename), caption="Camera 0", use_container_width=True)
            logger.info(f"Saved snapshot from Camera 0: {filename}")
            update_gallery()
    
    with cols[1]:
        cam1_img = st.camera_input("Take a picture - Cam 1", key="cam1")
        if cam1_img:
            filename = generate_filename("cam1")
            if st.session_state['cam1_file'] and os.path.exists(st.session_state['cam1_file']):
                os.remove(st.session_state['cam1_file'])
            with open(filename, "wb") as f:
                f.write(cam1_img.getvalue())
            st.session_state['cam1_file'] = filename
            entry = f"Snapshot from Cam 1: {filename}"
            if entry not in st.session_state['history']:
                st.session_state['history'] = [e for e in st.session_state['history'] if not e.startswith("Snapshot from Cam 1:")] + [entry]
            st.image(Image.open(filename), caption="Camera 1", use_container_width=True)
            logger.info(f"Saved snapshot from Camera 1: {filename}")
            update_gallery()

with tab_download:
    st.header("Download PDFs 📥")
    if st.button("Examples 📚"):
        example_urls = [
            "https://arxiv.org/pdf/2308.03892",
            "https://arxiv.org/pdf/1912.01703",
            "https://arxiv.org/pdf/2408.11039",
            "https://arxiv.org/pdf/2109.10282",
            "https://arxiv.org/pdf/2112.10752",
            "https://arxiv.org/pdf/2308.11236",
            "https://arxiv.org/pdf/1706.03762",
            "https://arxiv.org/pdf/2006.11239",
            "https://arxiv.org/pdf/2305.11207",
            "https://arxiv.org/pdf/2106.09685",
            "https://arxiv.org/pdf/2005.11401",
            "https://arxiv.org/pdf/2106.10504"
        ]
        st.session_state['pdf_urls'] = "\n".join(example_urls)
    
    url_input = st.text_area("Enter PDF URLs (one per line)", value=st.session_state.get('pdf_urls', ""), height=200)
    if st.button("Robo-Download 🤖"):
        urls = url_input.strip().split("\n")
        progress_bar = st.progress(0)
        status_text = st.empty()
        total_urls = len(urls)
        existing_pdfs = get_pdf_files()
        for idx, url in enumerate(urls):
            if url:
                output_path = pdf_url_to_filename(url)
                status_text.text(f"Fetching {idx + 1}/{total_urls}: {os.path.basename(output_path)}...")
                if output_path not in existing_pdfs:
                    if download_pdf(url, output_path):
                        st.session_state['downloaded_pdfs'][url] = output_path
                        logger.info(f"Downloaded PDF from {url} to {output_path}")
                        entry = f"Downloaded PDF: {output_path}"
                        if entry not in st.session_state['history']:
                            st.session_state['history'].append(entry)
                        st.session_state['asset_checkboxes'][output_path] = True
                    else:
                        st.error(f"Failed to nab {url} 😿")
                else:
                    st.info(f"Already got {os.path.basename(output_path)}! Skipping... 🐾")
                    st.session_state['downloaded_pdfs'][url] = output_path
                progress_bar.progress((idx + 1) / total_urls)
        status_text.text("Robo-Download complete! 🚀")
        update_gallery()
    
    mode = st.selectbox("Snapshot Mode", ["Single Page (High-Res)", "Two Pages (High-Res)", "All Pages (High-Res)"], key="download_mode")
    if st.button("Snapshot Selected 📸"):
        selected_pdfs = [path for path in get_gallery_files() if path.endswith('.pdf') and st.session_state['asset_checkboxes'].get(path, False)]
        if selected_pdfs:
            for pdf_path in selected_pdfs:
                if not os.path.exists(pdf_path):
                    st.warning(f"File not found: {pdf_path}. Skipping.")
                    continue
                mode_key = {"Single Page (High-Res)": "single", "Two Pages (High-Res)": "twopage", "All Pages (High-Res)": "allpages"}[mode]
                snapshots = asyncio.run(process_pdf_snapshot(pdf_path, mode_key))
                for snapshot in snapshots:
                    st.image(Image.open(snapshot), caption=snapshot, use_container_width=True)
                    st.session_state['asset_checkboxes'][snapshot] = True
            update_gallery()
        else:
            st.warning("No PDFs selected for snapshotting! Check some boxes in the sidebar.")

with tab_ocr:
    st.header("Test OCR 🔍")
    all_files = get_gallery_files()
    if all_files:
        if st.button("OCR All Assets 🚀"):
            full_text = "# OCR Results\n\n"
            for file in all_files:
                if file.endswith('.png'):
                    image = Image.open(file)
                else:
                    doc = fitz.open(file)
                    pix = doc[0].get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
                    image = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
                    doc.close()
                output_file = generate_filename(f"ocr_{os.path.basename(file)}", "txt")
                result = asyncio.run(process_ocr(image, output_file))
                full_text += f"## {os.path.basename(file)}\n\n{result}\n\n"
                entry = f"OCR Test: {file} -> {output_file}"
                if entry not in st.session_state['history']:
                    st.session_state['history'].append(entry)
            md_output_file = f"full_ocr_{int(time.time())}.md"
            with open(md_output_file, "w") as f:
                f.write(full_text)
            st.success(f"Full OCR saved to {md_output_file}")
            st.markdown(get_download_link(md_output_file, "text/markdown", "Download Full OCR Markdown"), unsafe_allow_html=True)
        selected_file = st.selectbox("Select Image or PDF", all_files, key="ocr_select")
        if selected_file:
            if selected_file.endswith('.png'):
                image = Image.open(selected_file)
            else:
                doc = fitz.open(selected_file)
                pix = doc[0].get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
                image = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
                doc.close()
            st.image(image, caption="Input Image", use_container_width=True)
            if st.button("Run OCR 🚀", key="ocr_run"):
                output_file = generate_filename("ocr_output", "txt")
                st.session_state['processing']['ocr'] = True
                result = asyncio.run(process_ocr(image, output_file))
                entry = f"OCR Test: {selected_file} -> {output_file}"
                if entry not in st.session_state['history']:
                    st.session_state['history'].append(entry)
                st.text_area("OCR Result", result, height=200, key="ocr_result")
                st.success(f"OCR output saved to {output_file}")
                st.session_state['processing']['ocr'] = False
            if selected_file.endswith('.pdf') and st.button("OCR All Pages 🚀", key="ocr_all_pages"):
                doc = fitz.open(selected_file)
                full_text = f"# OCR Results for {os.path.basename(selected_file)}\n\n"
                for i in range(len(doc)):
                    pix = doc[i].get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
                    image = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
                    output_file = generate_filename(f"ocr_page_{i}", "txt")
                    result = asyncio.run(process_ocr(image, output_file))
                    full_text += f"## Page {i + 1}\n\n{result}\n\n"
                    entry = f"OCR Test: {selected_file} Page {i + 1} -> {output_file}"
                    if entry not in st.session_state['history']:
                        st.session_state['history'].append(entry)
                md_output_file = f"full_ocr_{os.path.basename(selected_file)}_{int(time.time())}.md"
                with open(md_output_file, "w") as f:
                    f.write(full_text)
                st.success(f"Full OCR saved to {md_output_file}")
                st.markdown(get_download_link(md_output_file, "text/markdown", "Download Full OCR Markdown"), unsafe_allow_html=True)
    else:
        st.warning("No assets in gallery yet. Use Camera Snap or Download PDFs!")

with tab_build:
    st.header("Build Titan 🌱")
    model_type = st.selectbox("Model Type", ["Causal LM", "Diffusion"], key="build_type")
    base_model = st.selectbox(
        "Select Tiny Model",
        ["HuggingFaceTB/SmolLM-135M", "Qwen/Qwen1.5-0.5B-Chat"] if model_type == "Causal LM" 
        else ["OFA-Sys/small-stable-diffusion-v0", "stabilityai/stable-diffusion-2-base"]
    )
    model_name = st.text_input("Model Name", f"tiny-titan-{int(time.time())}")
    domain = st.text_input("Target Domain", "general")
    if st.button("Download Model ⬇️"):
        config = (ModelConfig if model_type == "Causal LM" else DiffusionConfig)(
            name=model_name, base_model=base_model, size="small", domain=domain
        )
        builder = ModelBuilder() if model_type == "Causal LM" else DiffusionBuilder()
        builder.load_model(base_model, config); builder.save_model(config.model_path)
        st.session_state['builder'] = builder; st.session_state['model_loaded'] = True
        st.session_state['selected_model_type'] = model_type; st.session_state['selected_model'] = config.model_path
        entry = f"Built {model_type} model: {model_name}"
        if entry not in st.session_state['history']:
            st.session_state['history'].append(entry)
        st.success(f"Model downloaded and saved to {config.model_path}! 🎉"); st.rerun()

with tab_imggen:
    st.header("Test Image Gen 🎨")
    all_files = get_gallery_files()
    if all_files:
        selected_file = st.selectbox("Select Image or PDF", all_files, key="gen_select")
        if selected_file:
            if selected_file.endswith('.png'):
                image = Image.open(selected_file)
            else:
                doc = fitz.open(selected_file)
                pix = doc[0].get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
                image = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
                doc.close()
            st.image(image, caption="Reference Image", use_container_width=True)
            prompt = st.text_area("Prompt", "Generate a neon superhero version of this image", key="gen_prompt")
            if st.button("Run Image Gen 🚀", key="gen_run"):
                output_file = generate_filename("gen_output", "png")
                st.session_state['processing']['gen'] = True
                result = asyncio.run(process_image_gen(prompt, output_file))
                entry = f"Image Gen Test: {prompt} -> {output_file}"
                if entry not in st.session_state['history']:
                    st.session_state['history'].append(entry)
                st.image(result, caption="Generated Image", use_container_width=True)
                st.success(f"Image saved to {output_file}")
                st.session_state['processing']['gen'] = False
    else:
        st.warning("No images or PDFs in gallery yet. Use Camera Snap or Download PDFs!")
    update_gallery()

with tab_pdf_process:
    st.header("PDF Process")
    st.subheader("Upload PDFs for GPT-based text extraction")
    gpt_models = ["gpt-4o", "gpt-4o-mini"]
    selected_gpt_model = st.selectbox("Select GPT Model", gpt_models, key="pdf_gpt_model")
    detail_level = st.selectbox("Detail Level", ["auto", "low", "high"], key="pdf_detail_level")
    uploaded_pdfs = st.file_uploader("Upload PDF files", type=["pdf"], accept_multiple_files=True, key="pdf_process_uploader")
    view_mode = st.selectbox("View Mode", ["Single Page", "Double Page"], key="pdf_view_mode")
    if st.button("Process Uploaded PDFs", key="process_pdfs"):
        combined_text = ""
        for pdf_file in uploaded_pdfs:
            pdf_bytes = pdf_file.read()
            temp_pdf_path = f"temp_{pdf_file.name}"
            with open(temp_pdf_path, "wb") as f:
                f.write(pdf_bytes)
            try:
                doc = fitz.open(temp_pdf_path)
                st.write(f"Processing {pdf_file.name} with {len(doc)} pages")
                if view_mode == "Single Page":
                    for i, page in enumerate(doc):
                        pix = page.get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
                        img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
                        st.image(img, caption=f"{pdf_file.name} Page {i+1}")
                        gpt_text = process_image_with_prompt(img, "Extract the electronic text from image", model=selected_gpt_model, detail=detail_level)
                        combined_text += f"\n## {pdf_file.name} - Page {i+1}\n\n{gpt_text}\n"
                else:
                    pages = list(doc)
                    for i in range(0, len(pages), 2):
                        if i+1 < len(pages):
                            pix1 = pages[i].get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
                            img1 = Image.frombytes("RGB", [pix1.width, pix1.height], pix1.samples)
                            pix2 = pages[i+1].get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
                            img2 = Image.frombytes("RGB", [pix2.width, pix2.height], pix2.samples)
                            total_width = img1.width + img2.width
                            max_height = max(img1.height, img2.height)
                            combined_img = Image.new("RGB", (total_width, max_height))
                            combined_img.paste(img1, (0, 0))
                            combined_img.paste(img2, (img1.width, 0))
                            st.image(combined_img, caption=f"{pdf_file.name} Pages {i+1}-{i+2}")
                            gpt_text = process_image_with_prompt(combined_img, "Extract the electronic text from image", model=selected_gpt_model, detail=detail_level)
                            combined_text += f"\n## {pdf_file.name} - Pages {i+1}-{i+2}\n\n{gpt_text}\n"
                        else:
                            pix = pages[i].get_pixmap(matrix=fitz.Matrix(2.0, 2.0))
                            img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
                            st.image(img, caption=f"{pdf_file.name} Page {i+1}")
                            gpt_text = process_image_with_prompt(img, "Extract the electronic text from image", model=selected_gpt_model, detail=detail_level)
                            combined_text += f"\n## {pdf_file.name} - Page {i+1}\n\n{gpt_text}\n"
                doc.close()
            except Exception as e:
                st.error(f"Error processing {pdf_file.name}: {str(e)}")
            finally:
                os.remove(temp_pdf_path)
        output_filename = generate_filename("processed_pdf", "md")
        with open(output_filename, "w", encoding="utf-8") as f:
            f.write(combined_text)
        st.success(f"PDF processing complete. MD file saved as {output_filename}")
        st.markdown(get_download_link(output_filename, "text/markdown", "Download Processed PDF MD"), unsafe_allow_html=True)

with tab_image_process:
    st.header("Image Process")
    st.subheader("Upload Images for GPT-based OCR")
    gpt_models = ["gpt-4o", "gpt-4o-mini"]
    selected_gpt_model = st.selectbox("Select GPT Model", gpt_models, key="img_gpt_model")
    detail_level = st.selectbox("Detail Level", ["auto", "low", "high"], key="img_detail_level")
    prompt_img = st.text_input("Enter prompt for image processing", "Extract the electronic text from image", key="img_process_prompt")
    uploaded_images = st.file_uploader("Upload image files", type=["png", "jpg", "jpeg"], accept_multiple_files=True, key="image_process_uploader")
    if st.button("Process Uploaded Images", key="process_images"):
        combined_text = ""
        for img_file in uploaded_images:
            try:
                img = Image.open(img_file)
                st.image(img, caption=img_file.name)
                gpt_text = process_image_with_prompt(img, prompt_img, model=selected_gpt_model, detail=detail_level)
                combined_text += f"\n## {img_file.name}\n\n{gpt_text}\n"
            except Exception as e:
                st.error(f"Error processing image {img_file.name}: {str(e)}")
        output_filename = generate_filename("processed_image", "md")
        with open(output_filename, "w", encoding="utf-8") as f:
            f.write(combined_text)
        st.success(f"Image processing complete. MD file saved as {output_filename}")
        st.markdown(get_download_link(output_filename, "text/markdown", "Download Processed Image MD"), unsafe_allow_html=True)

with tab_md_gallery:
    st.header("MD Gallery and GPT Processing")
    gpt_models = ["gpt-4o", "gpt-4o-mini"]
    selected_gpt_model = st.selectbox("Select GPT Model", gpt_models, key="md_gpt_model")
    md_files = sorted(glob.glob("*.md"))
    if md_files:
        st.subheader("Individual File Processing")
        cols = st.columns(2)
        for idx, md_file in enumerate(md_files):
            with cols[idx % 2]:
                st.write(md_file)
                if st.button(f"Process {md_file}", key=f"process_md_{md_file}"):
                    try:
                        with open(md_file, "r", encoding="utf-8") as f:
                            content = f.read()
                        prompt_md = "Summarize this into markdown outline with emojis and number the topics 1..12"
                        result_text = process_text_with_prompt(content, prompt_md, model=selected_gpt_model)
                        st.markdown(result_text)
                        output_filename = generate_filename(f"processed_{os.path.splitext(md_file)[0]}", "md")
                        with open(output_filename, "w", encoding="utf-8") as f:
                            f.write(result_text)
                        st.markdown(get_download_link(output_filename, "text/markdown", f"Download {output_filename}"), unsafe_allow_html=True)
                    except Exception as e:
                        st.error(f"Error processing {md_file}: {str(e)}")
        st.subheader("Batch Processing")
        st.write("Select MD files to combine and process:")
        selected_md = {}
        for md_file in md_files:
            selected_md[md_file] = st.checkbox(md_file, key=f"checkbox_md_{md_file}")
        batch_prompt = st.text_input("Enter batch processing prompt", "Summarize this into markdown outline with emojis and number the topics 1..12", key="batch_prompt")
        if st.button("Process Selected MD Files", key="process_batch_md"):
            combined_content = ""
            for md_file, selected in selected_md.items():
                if selected:
                    try:
                        with open(md_file, "r", encoding="utf-8") as f:
                            combined_content += f"\n## {md_file}\n" + f.read() + "\n"
                    except Exception as e:
                        st.error(f"Error reading {md_file}: {str(e)}")
            if combined_content:
                result_text = process_text_with_prompt(combined_content, batch_prompt, model=selected_gpt_model)
                st.markdown(result_text)
                output_filename = generate_filename("batch_processed_md", "md")
                with open(output_filename, "w", encoding="utf-8") as f:
                    f.write(result_text)
                st.success(f"Batch processing complete. MD file saved as {output_filename}")
                st.markdown(get_download_link(output_filename, "text/markdown", "Download Batch Processed MD"), unsafe_allow_html=True)
            else:
                st.warning("No MD files selected.")
    else:
        st.warning("No MD files found.")