Fix cache_dir crash, modernize UI, switch to Groq

.streamlit/config.toml

@@ -0,0 +1,13 @@
+[theme]
+base = "light"
+primaryColor = "#6366f1"
+backgroundColor = "#f5f7ff"
+secondaryBackgroundColor = "#ffffff"
+textColor = "#1e293b"
+font = "sans serif"
+
+[server]
+headless = true
+
+[browser]
+gatherUsageStats = false

Dockerfile

@@ -9,12 +9,13 @@ RUN apt-get update && apt-get install -y \
     && rm -rf /var/lib/apt/lists/*
 
 COPY requirements.txt ./
+COPY .streamlit/ .streamlit/
 COPY src/ ./src/
 
-RUN pip3 install -r requirements.txt
+RUN pip3 install --no-cache-dir -r requirements.txt
 
 EXPOSE 8501
 
 HEALTHCHECK CMD curl --fail http://localhost:8501/_stcore/health
 
-ENTRYPOINT ["streamlit", "run", "src/streamlit_app.py", "--server.port=8501", "--server.address=0.0.0.0"]
+ENTRYPOINT ["streamlit", "run", "src/streamlit_app.py", "--server.port=8501", "--server.address=0.0.0.0"]

README.md

@@ -1,55 +1,64 @@
 ---
 title: Credo AI
 emoji: 🚀
-colorFrom: red
-colorTo: red
+colorFrom: indigo
+colorTo: purple
 sdk: docker
 app_port: 8501
-tags:
-- streamlit
 pinned: false
 short_description: The Two-Brain Misinformation Detector
 license: mit
 ---
-Credo AI - The Two-Brain Detector emoji: 🧠⚡️ colorFrom: blue colorTo: purple sdk: streamlit sdk_version: 1.33.0 app_file: app.py pinned: false
-🧠 Credo AI: The Two-Brain Misinformation Detector
-Welcome to the live demo for our Hack2Skill Hackathon project!
 
-This application is a proof-of-concept powered by a unique "Two-Brain" AI system designed to provide both rapid and in-depth analysis of news articles and other text. It demonstrates a powerful new architecture for tackling misinformation.
+# 🧠 Credo AI: The Two-Brain Misinformation Detector
 
-Key Features
-Live URL & Text Analysis: Paste any text or a URL to a news article to get an instant analysis.
+Welcome to the live demo of **Credo AI** — a "Two-Brain" AI system that provides rapid and in-depth analysis of news articles and text, built for the Hack2Skill Hackathon 2025.
 
-Dual-AI Verdict: Get a fast, high-confidence FAKE/REAL verdict from our specialist model, alongside a nuanced 6-point breakdown from our expert model.
+**Live Space:** https://huggingface.co/spaces/Arko007/Credo_AI
 
-Gemini-Powered Summaries: Receive a clear, conversational summary of the findings, powered by Google's Gemini API.
+## Key Features
 
-Analysis History: All of your queries are saved in your session and can be reviewed on the "Analysis History" page.
+- **Live URL & Text Analysis:** Paste any text or a URL to a news article for an instant analysis.
+- **Dual-AI Verdict:** A fast, high-confidence FAKE/REAL verdict from a specialist model, plus a nuanced check from an expert model.
+- **Groq-Powered Explanations:** Clear, conversational explanations of the findings, powered by Groq LPU inference (`qwen/qwen3.6-27b`, fallback `openai/gpt-oss-120b`).
+- **Web Cross-Reference:** Optional Tavily-based search to cross-check claims against live sources.
+- **Analysis History:** All queries are saved in your session and reviewable on the History page.
+- **Export:** Download each analysis as a JSON report.
 
-⚠️ Important Limitations & Prototype Nature
-This is a hackathon prototype, and while powerful, it has important limitations that highlight key challenges in the field of AI fact-checking.
+## Models
 
-1. Pattern Recognition vs. World Knowledge
-Our AI models are expert pattern detectors, not true fact-checkers with a database of world knowledge. They have been trained to identify the linguistic style, tone, and structure of fake news with incredible accuracy.
+| Brain | Model | Base | Role |
+|---|---|---|---|
+| Brain 1 | `Arko007/fake-news-liar-political` | RoBERTa-base | Political (US-centric) FAKE/REAL |
+| Brain 2 | `Arko007/fact-check1-v3-final` | DeBERTa-v3-large | General FAKE/REAL specialist |
 
-However, this means the system can be fooled by simple, declarative statements that are factually incorrect but do not fit the stylistic patterns of fake news.
+The app automatically routes political text to Brain 1 and everything else to Brain 2.
 
-For example, the model may incorrectly classify "The sun rises in the West" as REAL.
+## Setup
 
-This happens because the statement, while false, is simple and declarative. It doesn't contain the sensationalism, complex grammar, or emotional language that the AI has learned to associate with fake news. This is a classic "common sense" problem in AI and demonstrates that our model is a sophisticated misinformation detector, not a universal truth engine.
+### Secrets (Space Settings → Variables and Secrets)
 
-2. Limited Training Data & Domain Bias
-As a prototype built within a limited timeframe, the models were trained on a relatively small amount of publicly available data. This data was primarily focused on Western political news. As a result, the model's accuracy will be significantly lower on:
+- `GROQ_API_KEY` — required for AI-powered explanations (https://console.groq.com)
+- `TAVILY_API_KEY` — optional, enables live web cross-referencing (https://tavily.com)
 
-Topics outside of politics (e.g., science, finance, health).
+The app works in Basic Mode without any keys, using built-in summaries and fallback analysis.
 
-News from different geopolitical contexts (e.g., Indian or Asian news).
+## ⚠️ Important Limitations
 
-3. Future Work
-These limitations highlight the path forward. The next version of Credo AI would involve:
+This is a hackathon prototype. The models are expert **pattern detectors**, not universal truth engines:
 
-Integrating a Knowledge Graph: Connecting the AI to a database of real-world facts to overcome the common sense problem.
+1. **Pattern Recognition vs. World Knowledge** — simple, declarative but factually false statements (e.g., "The sun rises in the West") can be missed because they don't match fake-news stylistic patterns.
+2. **Limited Training Data & Domain Bias** — models were trained mostly on Western political news; accuracy drops on science, finance, health, and non-Western news contexts.
+3. **Future Work** — knowledge-graph grounding, multi-domain training, and global datasets.
 
-Training on a Larger, Multi-Domain Dataset: Expanding the training data to include millions of articles from diverse topics and regions to create a truly global tool.
+Always treat results as assistive, not authoritative.
 
-Built by the Data Dragons 🐉 for the Hack2Skill Hackathon 2025.
+## Tech Stack
+
+- **Frontend:** Streamlit + custom CSS
+- **Models:** Hugging Face Transformers (PyTorch), RoBERTa & DeBERTa-v3
+- **Explanations:** Groq Cloud API
+- **Search:** Tavily
+- **Scraping:** Beautiful Soup + lxml
+
+Built by the Data Dragons 🐉 for the Hack2Skill Hackathon 2025.

requirements.txt

@@ -1,11 +1,11 @@
-streamlit>=1.28.0
-transformers>=4.40.0
+streamlit>=1.46.0
+transformers>=4.57.0
 torch>=2.5.0
-pandas>=2.0.0
-plotly>=5.15.0
-scikit-learn>=1.3.0
-requests>=2.31.0
+groq>=0.14.0
+tavily-python>=0.5.0
+pandas>=2.2.0
+numpy>=1.26.0
+requests>=2.32.0
 beautifulsoup4>=4.12.0
-google-generativeai>=0.3.0
-numpy>=1.24.0
-sentencepiece>=0.1.99
+lxml>=5.2.0
+sentencepiece>=0.2.0

src/streamlit_app.py

@@ -10,41 +10,297 @@ import streamlit as st
 import torch
 from transformers import pipeline
 
-# Import google-generativeai with fallback
 try:
-    import google.generativeai as genai
-    GENAI_AVAILABLE = True
-except ImportError:
-    GENAI_AVAILABLE = False
+    from groq import Groq
+    GROQ_CLIENT = Groq(api_key=os.getenv("GROQ_API_KEY")) if os.getenv("GROQ_API_KEY") else None
+    GROQ_AVAILABLE = GROQ_CLIENT is not None
+except Exception:
+    GROQ_CLIENT = None
+    GROQ_AVAILABLE = False
 
 try:
     from tavily import TavilyClient
-    TAVILY_CLIENT = TavilyClient(api_key=os.getenv("TAVILY_API_KEY"))
-    TAVILY_AVAILABLE = True
+    TAVILY_CLIENT = TavilyClient(api_key=os.getenv("TAVILY_API_KEY")) if os.getenv("TAVILY_API_KEY") else None
+    TAVILY_AVAILABLE = TAVILY_CLIENT is not None
 except Exception:
+    TAVILY_CLIENT = None
     TAVILY_AVAILABLE = False
 
-# Environment and Cache Setup
-os.environ['HF_HOME'] = '/tmp'
-os.environ['TRANSFORMERS_CACHE'] = '/tmp'
-os.environ['HF_HUB_CACHE'] = '/tmp'
-
-# Model IDs
 BRAIN_1_MODEL = "Arko007/fake-news-liar-political"
 BRAIN_2_MODEL = "Arko007/fact-check1-v3-final"
 
-# Streamlit config and styling (full CSS as you provided earlier)
+GROQ_MODELS = [
+    "qwen/qwen3.6-27b",
+    "openai/gpt-oss-120b",
+    "openai/gpt-oss-20b",
+]
+
 st.set_page_config(
     page_title="Credo AI | Truth Detection Platform",
     page_icon="🧠",
     layout="wide",
-    initial_sidebar_state="expanded"
+    initial_sidebar_state="expanded",
 )
 
 st.markdown("""
 <style>
-/* All your full CSS styling here, unchanged */
-[...your full CSS from before...]
+@import url('https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700;800&display=swap');
+
+html, body, [class*="css"], [class*="st-"], .stApp {
+    font-family: 'Inter', -apple-system, BlinkMacSystemFont, 'Segoe UI', sans-serif;
+}
+
+.stApp {
+    background: linear-gradient(135deg, #f5f7ff 0%, #eef1fb 50%, #e9eefa 100%);
+    color: #1e293b;
+}
+
+h1, h2, h3, h4 {
+    color: #1e293b;
+    letter-spacing: -0.01em;
+}
+
+[data-testid="stSidebar"] {
+    background: #ffffff;
+    border-right: 1px solid #e5e9f2;
+}
+[data-testid="stSidebar"] .stRadio label {
+    color: #475569;
+}
+
+.hero-container {
+    background: linear-gradient(135deg, #ffffff 0%, #f8faff 100%);
+    border: 1px solid #e5e9f2;
+    border-radius: 20px;
+    padding: 2.5rem 2rem;
+    text-align: center;
+    box-shadow: 0 10px 30px rgba(99, 102, 241, 0.08);
+    margin-bottom: 1.5rem;
+}
+.main-title {
+    background: linear-gradient(90deg, #4f46e5, #7c3aed, #4f46e5);
+    background-size: 200% auto;
+    -webkit-background-clip: text;
+    -webkit-text-fill-color: transparent;
+    background-clip: text;
+    animation: shimmer 6s linear infinite;
+    font-size: 2.6rem;
+    font-weight: 800;
+    letter-spacing: -0.02em;
+    margin: 0 0 0.75rem 0;
+}
+@keyframes shimmer {
+    to { background-position: 200% center; }
+}
+.hero-subtitle {
+    color: #64748b;
+    font-size: 1.1rem;
+    max-width: 760px;
+    margin: 0 auto 1.5rem auto;
+    line-height: 1.7;
+}
+.metrics-container {
+    display: flex;
+    gap: 1rem;
+    justify-content: center;
+    flex-wrap: wrap;
+}
+.metric-card {
+    background: #ffffff;
+    border: 1px solid #e5e9f2;
+    border-radius: 16px;
+    padding: 1rem 1.75rem;
+    min-width: 130px;
+    box-shadow: 0 6px 16px rgba(99, 102, 241, 0.06);
+    transition: transform 0.2s ease, box-shadow 0.2s ease;
+}
+.metric-card:hover {
+    transform: translateY(-3px);
+    box-shadow: 0 10px 24px rgba(99, 102, 241, 0.14);
+}
+.metric-value {
+    display: block;
+    font-size: 1.9rem;
+    font-weight: 800;
+    background: linear-gradient(90deg, #4f46e5, #7c3aed);
+    -webkit-background-clip: text;
+    -webkit-text-fill-color: transparent;
+    background-clip: text;
+}
+.metric-label {
+    display: block;
+    color: #94a3b8;
+    font-size: 0.78rem;
+    text-transform: uppercase;
+    letter-spacing: 0.08em;
+    margin-top: 0.2rem;
+}
+
+.summary-box {
+    background: #ffffff;
+    border: 1px solid #e5e9f2;
+    border-left: 5px solid #6366f1;
+    border-radius: 14px;
+    padding: 1.25rem 1.5rem;
+    color: #334155;
+    line-height: 1.7;
+    font-size: 1.02rem;
+    box-shadow: 0 6px 18px rgba(99, 102, 241, 0.06);
+}
+
+.verdict-container {
+    border-radius: 18px;
+    padding: 2rem 1.5rem;
+    text-align: center;
+    box-shadow: 0 10px 26px rgba(15, 23, 42, 0.12);
+}
+.verdict-fake {
+    background: linear-gradient(135deg, #ef4444, #dc2626);
+}
+.verdict-real {
+    background: linear-gradient(135deg, #10b981, #059669);
+}
+.verdict-uncertain {
+    background: linear-gradient(135deg, #f59e0b, #d97706);
+}
+.verdict-text {
+    font-size: 2.2rem;
+    font-weight: 800;
+    letter-spacing: 0.12em;
+    color: #ffffff;
+    text-shadow: 0 2px 8px rgba(0, 0, 0, 0.18);
+}
+
+.glass-card {
+    background: #ffffff;
+    border: 1px solid #e5e9f2;
+    border-radius: 18px;
+    padding: 1.75rem;
+    box-shadow: 0 8px 24px rgba(99, 102, 241, 0.07);
+}
+
+.footer-enhanced {
+    margin-top: 2.5rem;
+    padding: 1.5rem 2rem;
+    background: #ffffff;
+    border: 1px solid #e5e9f2;
+    border-radius: 16px;
+    text-align: center;
+    color: #64748b;
+    box-shadow: 0 6px 18px rgba(99, 102, 241, 0.06);
+}
+.footer-features {
+    display: flex;
+    gap: 1.25rem;
+    justify-content: center;
+    flex-wrap: wrap;
+    margin-bottom: 0.75rem;
+}
+.footer-feature {
+    display: flex;
+    align-items: center;
+    gap: 0.4rem;
+    font-size: 0.9rem;
+    font-weight: 600;
+    color: #475569;
+}
+.footer-feature-icon {
+    font-size: 1.1rem;
+}
+
+.stButton > button, .stDownloadButton > button {
+    border-radius: 12px;
+    border: none;
+    font-weight: 600;
+    transition: all 0.2s ease;
+}
+.stButton > button[kind="primary"], .stDownloadButton > button {
+    background: linear-gradient(90deg, #4f46e5, #7c3aed);
+    color: #ffffff;
+    box-shadow: 0 6px 16px rgba(99, 102, 241, 0.3);
+}
+.stButton > button[kind="primary"]:hover, .stDownloadButton > button:hover {
+    transform: translateY(-2px);
+    box-shadow: 0 10px 22px rgba(99, 102, 241, 0.42);
+}
+.stButton > button[kind="secondary"] {
+    background: #ffffff;
+    color: #4f46e5;
+    border: 1px solid #d5daf0;
+}
+.stButton > button[kind="secondary"]:hover {
+    border-color: #6366f1;
+    background: #f5f6ff;
+    transform: translateY(-2px);
+}
+.stButton > button:disabled, .stDownloadButton > button:disabled {
+    opacity: 0.55;
+    box-shadow: none;
+}
+
+.stTextInput input, .stTextArea textarea, [data-baseweb="select"] > div {
+    border-radius: 12px !important;
+    border: 1px solid #dde3f0 !important;
+    background: #ffffff !important;
+}
+.stTextInput input:focus, .stTextArea textarea:focus, [data-baseweb="select"] > div:focus-within {
+    border-color: #6366f1 !important;
+    box-shadow: 0 0 0 3px rgba(99, 102, 241, 0.15) !important;
+}
+
+[data-testid="stMetric"] {
+    background: #ffffff;
+    border: 1px solid #e5e9f2;
+    border-radius: 14px;
+    padding: 1rem 1.25rem;
+    box-shadow: 0 6px 16px rgba(99, 102, 241, 0.06);
+}
+[data-testid="stMetricValue"] {
+    font-weight: 800;
+    color: #4f46e5;
+}
+
+.stTabs [data-baseweb="tab"] {
+    border-radius: 10px;
+}
+.stTabs [data-baseweb="tab-highlight"], .stTabs [aria-selected="true"] {
+    background: #eef0ff !important;
+    color: #4f46e5 !important;
+    border-radius: 10px !important;
+}
+
+[data-testid="stExpander"] {
+    background: #ffffff;
+    border: 1px solid #e5e9f2 !important;
+    border-radius: 14px !important;
+    box-shadow: 0 6px 16px rgba(99, 102, 241, 0.05);
+}
+
+[data-testid="stAlert"] {
+    border-radius: 12px;
+    border: none;
+}
+
+[data-testid="stFileUploader"] section {
+    border: 1px dashed #c7cdf0;
+    border-radius: 12px;
+    background: #fafbff;
+}
+
+[data-testid="stStatusWidget"] {
+    background: #ffffff;
+    border: 1px solid #e5e9f2;
+    border-radius: 14px;
+    box-shadow: 0 6px 16px rgba(99, 102, 241, 0.06);
+}
+
+.stProgress > div > div > div {
+    background: linear-gradient(90deg, #4f46e5, #7c3aed);
+}
+
+a {
+    color: #6366f1;
+}
 </style>
 """, unsafe_allow_html=True)
 
@@ -57,17 +313,16 @@ def load_ai_models():
             classifier_b1 = pipeline(
                 "text-classification",
                 model=BRAIN_1_MODEL,
+                tokenizer=BRAIN_1_MODEL,
                 return_all_scores=False,
                 device=0 if torch.cuda.is_available() else -1,
-                tokenizer=BRAIN_1_MODEL,
-                cache_dir='/tmp/huggingface_cache'
             )
             st.write("🎯 Initializing Brain 2 (General)...")
             classifier_b2 = pipeline(
                 "text-classification",
                 model=BRAIN_2_MODEL,
+                return_all_scores=False,
                 device=0 if torch.cuda.is_available() else -1,
-                cache_dir='/tmp/huggingface_cache'
             )
             status.update(label="✅ AI models loaded successfully!", state="complete")
             return classifier_b1, classifier_b2
@@ -76,6 +331,28 @@ def load_ai_models():
         return None, None
 
 
+def normalize_brain1_label(raw_label):
+    label = str(raw_label).upper()
+    if label in ("FAKE", "REAL"):
+        return label
+    if label == "LABEL_0":
+        return "FAKE"
+    if label == "LABEL_1":
+        return "REAL"
+    return label
+
+
+def normalize_brain2_label(raw_label):
+    label = str(raw_label).upper()
+    if label in ("FAKE", "REAL"):
+        return label
+    if label == "LABEL_1":
+        return "FAKE"
+    if label == "LABEL_0":
+        return "REAL"
+    return label
+
+
 def tavily_search(query):
     if not TAVILY_AVAILABLE:
         return None
@@ -95,64 +372,81 @@ def is_us_political(text):
     keywords = [
         "president", "congress", "senate", "house", "democrat", "republican",
         "biden", "trump", "politics", "political", "us government", "white house",
-        "politi", "liar", "election", "campaign", "supreme court"
+        "politi", "liar", "election", "campaign", "supreme court",
     ]
     text_lower = text.lower()
     return any(kw in text_lower for kw in keywords)
 
 
-def generate_gemini_explanation(text, classification, confidence):
-    try:
-        prompt = (
-            f"Analyze this content classified as {classification} (confidence approx {confidence:.1f}%).\n"
-            f"Content: {text[:400]}...\n"
-            f"Provide a concise professional explanation of why this classification is correct or not.\n"
-            f"If the classification appears incorrect based on real-time facts, correct it and explain."
-        )
-        model = genai.GenerativeModel(model_name="gemini-2.0-flash")
-        response = model.generate_content(prompt)
-        return response.text
-    except Exception:
-        return f"Content classified as {classification} with confidence {confidence:.1f}%. Explanation unavailable."
+def generate_ai_explanation(text, classification, confidence):
+    if not GROQ_AVAILABLE:
+        return None
+    prompt = (
+        f"Analyze this content that an AI model classified as '{classification}' "
+        f"with {confidence:.1f}% confidence.\n\n"
+        f"Content: {text[:400]}...\n\n"
+        f"Provide a concise professional explanation (2-4 sentences) of why this "
+        f"classification is or isn't correct. If the classification appears wrong, "
+        f"say so explicitly and explain."
+    )
+    for model in GROQ_MODELS:
+        try:
+            response = GROQ_CLIENT.chat.completions.create(
+                model=model,
+                messages=[{"role": "user", "content": prompt}],
+                temperature=0.3,
+                max_tokens=400,
+            )
+            content = response.choices[0].message.content
+            if content and content.strip():
+                return content.strip()
+        except Exception:
+            continue
+    return None
 
 
 def analyze_with_models(text, classifier_b1, classifier_b2):
     text_stripped = text.strip()
     use_brain1 = is_us_political(text_stripped)
+    brain_name = "Brain 1 (Political)" if use_brain1 else "Brain 2 (General)"
+    classifier = classifier_b1 if use_brain1 else classifier_b2
+
+    try:
+        results = classifier(text_stripped, truncation=True)
+        raw_label = results[0]["label"]
+        if use_brain1:
+            label = normalize_brain1_label(raw_label)
+        else:
+            label = normalize_brain2_label(raw_label)
+        confidence = float(results[0]["score"]) * 100.0
+    except Exception:
+        label, confidence, summary = get_fallback_analysis(text_stripped)
+        return label, confidence, summary, brain_name, False
+
+    summary, corrected = build_summary(text_stripped, label, confidence)
+    return corrected, confidence, summary, brain_name, True
 
-    if use_brain1:
-        results = classifier_b1(text_stripped)
-    else:
-        results = classifier_b2(text_stripped)
 
-    label = results[0]['label']
-    confidence = random.uniform(85.0, 99.5)
+def build_summary(text, label, confidence):
+    ai_explanation = generate_ai_explanation(text, label, confidence)
+    if ai_explanation:
+        lowered = ai_explanation.lower()
+        markers = ["incorrect", "wrong", f"not {label.lower()}", "misclassification"]
+        if any(marker in lowered for marker in markers):
+            corrected = "REAL" if label == "FAKE" else "FAKE"
+            return ai_explanation, corrected
+        return ai_explanation, label
 
     if TAVILY_AVAILABLE:
-        tavily_info = tavily_search(text_stripped)
+        tavily_info = tavily_search(text)
         if tavily_info:
-            if GENAI_AVAILABLE and API_CONFIGURED:
-                gemini_output = generate_gemini_explanation(text_stripped, label, confidence)
-                gem_label = label
-                if (
-                    "incorrect" in gemini_output.lower() or
-                    ("not " + label.lower()) in gemini_output.lower() or
-                    ("wrong" in gemini_output.lower())
-                ):
-                    gem_label = "REAL" if label == "FAKE" else "FAKE"
-                    label = gem_label
-                summary = gemini_output
-            else:
-                summary = f"Content classified as {label} by model. Confidence: {confidence:.1f}%."
-        else:
-            summary = f"Content classified as {label} by model. Confidence: {confidence:.1f}%."
-    else:
-        if GENAI_AVAILABLE and API_CONFIGURED:
-            summary = generate_gemini_explanation(text_stripped, label, confidence)
-        else:
-            summary = f"Content classified as {label} by model. Confidence: {confidence:.1f}%."
-
-    return label, confidence, summary
+            return (
+                f"Content classified as {label} by the model with {confidence:.1f}% "
+                f"confidence, cross-referenced against live web sources.", label,
+            )
+    return (
+        f"Content classified as {label} by the model with {confidence:.1f}% confidence.", label,
+    )
 
 
 def get_fallback_analysis(text):
@@ -162,31 +456,45 @@ def get_fallback_analysis(text):
     fake_score = sum(1 for word in fake_indicators if word in text_lower)
     real_score = sum(1 for word in real_indicators if word in text_lower)
     if fake_score > real_score:
-        return "FAKE", random.uniform(85.0, 99.5), "Fallback heuristic analysis: Likely FAKE content detected."
-    elif real_score > fake_score:
-        return "REAL", random.uniform(85.0, 99.5), "Fallback heuristic analysis: Likely REAL content detected."
-    else:
-        return "UNCERTAIN", random.uniform(85.0, 99.5), "Fallback heuristic analysis: Unable to classify definitively."
+        return "FAKE", random.uniform(60.0, 80.0), "Fallback heuristic analysis: Likely FAKE content detected."
+    if real_score > fake_score:
+        return "REAL", random.uniform(60.0, 80.0), "Fallback heuristic analysis: Likely REAL content detected."
+    return "UNCERTAIN", 50.0, "Fallback heuristic analysis: Unable to classify definitively."
 
 
 @st.cache_data(show_spinner=False, ttl=300)
 def fetch_web_content(url):
     try:
-        headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 Chrome/91.0.4472.124 Safari/537.36'}
+        headers = {
+            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 '
+                          'Chrome/124.0.0.0 Safari/537.36'
+        }
         response = requests.get(url, headers=headers, timeout=15)
         response.raise_for_status()
         soup = BeautifulSoup(response.content, 'html.parser')
 
         for element in soup(['script', 'style', 'nav', 'footer', 'aside']):
             element.decompose()
-        title = soup.find('title')
-        title = title.get_text(strip=True) if title else "No title found"
+
+        title_tag = soup.find('title')
+        title = title_tag.get_text(strip=True) if title_tag else "No title found"
 
         paragraphs = soup.find_all('p')
-        content = " ".join([p.get_text(strip=True) for p in paragraphs if len(p.get_text(strip=True)) > 20])
+        content = " ".join(
+            p.get_text(strip=True)
+            for p in paragraphs
+            if len(p.get_text(strip=True)) > 20
+        )
 
         full_text = f"{title}\n\n{content}"
-        return {'success': True, 'title': title, 'content': content, 'full_text': full_text, 'word_count': len(full_text.split()), 'url': url}
+        return {
+            'success': True,
+            'title': title,
+            'content': content,
+            'full_text': full_text,
+            'word_count': len(full_text.split()),
+            'url': url,
+        }
     except Exception as e:
         return {'success': False, 'error': str(e)}
 
@@ -210,7 +518,9 @@ def process_analysis(user_input, input_method, classifier_b1, classifier_b2):
             text_to_analyze = text_to_analyze[:3000]
             st.write("✂️ Text truncated for optimal processing")
 
-        label, confidence, summary = analyze_with_models(text_to_analyze, classifier_b1, classifier_b2)
+        label, confidence, summary, brain_name, model_used = analyze_with_models(
+            text_to_analyze, classifier_b1, classifier_b2
+        )
 
         analysis_time = time.time() - start_time
         status.update(label="✅ Analysis complete!", state="complete")
@@ -220,8 +530,10 @@ def process_analysis(user_input, input_method, classifier_b1, classifier_b2):
         'confidence': confidence,
         'summary': summary,
         'analysis_time': analysis_time,
+        'brain': brain_name,
+        'model_used': model_used,
         'input': user_input[:200] + "..." if len(user_input) > 200 else user_input,
-        'full_input': user_input
+        'full_input': user_input,
     }
 
     st.session_state.current_results = results
@@ -241,7 +553,7 @@ def render_analysis_interface(classifier_b1, classifier_b2):
     input_method = st.selectbox(
         "Select input method:",
         ["Direct Text", "URL/Website", "File Upload"],
-        help="Choose how you want to provide content for fact-checking"
+        help="Choose how you want to provide content for fact-checking",
     )
     user_input = ""
     if input_method == "Direct Text":
@@ -249,13 +561,13 @@ def render_analysis_interface(classifier_b1, classifier_b2):
             "Enter text to analyze:",
             height=150,
             placeholder="Paste the content you want to fact-check here...",
-            help="Enter any text content for misinformation detection"
+            help="Enter any text content for misinformation detection",
         )
     elif input_method == "URL/Website":
         user_input = st.text_input(
             "Enter website URL:",
             placeholder="https://example.com/article",
-            help="Provide the URL of an article or webpage to analyze"
+            help="Provide the URL of an article or webpage to analyze",
         )
         if user_input and not user_input.startswith(('http://', 'https://')):
             st.warning("⚠️ Please enter a complete URL starting with http:// or https://")
@@ -263,17 +575,20 @@ def render_analysis_interface(classifier_b1, classifier_b2):
         uploaded_file = st.file_uploader(
             "Upload text file:",
             type=['txt', 'md'],
-            help="Upload a text file containing the content to analyze"
+            help="Upload a text file containing the content to analyze",
         )
         if uploaded_file:
             try:
-                user_input = str(uploaded_file.read(), "utf-8")
+                user_input = uploaded_file.getvalue().decode("utf-8", errors="replace")
                 st.success(f"✅ File loaded: {len(user_input)} characters")
                 if len(user_input) > 500:
-                    st.text_area("Content preview:", user_input[:500] + "...", height=100, disabled=True)
+                    st.text_area(
+                        "Content preview:", user_input[:500] + "...", height=100, disabled=True
+                    )
             except Exception as e:
                 st.error(f"❌ Error reading file: {str(e)}")
                 user_input = ""
+
     st.markdown("---")
     col1, col2, col3 = st.columns([3, 1, 1])
     with col1:
@@ -281,7 +596,7 @@ def render_analysis_interface(classifier_b1, classifier_b2):
             "🧠 Analyze with Dual-AI",
             type="primary",
             disabled=not user_input.strip(),
-            help="Start the AI-powered fact-checking analysis"
+            help="Start the AI-powered fact-checking analysis",
         )
     with col2:
         if st.button("🔄 Clear", help="Clear current results and start over"):
@@ -293,6 +608,7 @@ def render_analysis_interface(classifier_b1, classifier_b2):
         if st.button("📄 Export", disabled=not export_enabled, help="Export analysis results"):
             if export_enabled:
                 export_results()
+
     if analyze_btn:
         if not user_input.strip():
             st.warning("⚠️ Please provide some content to analyze.")
@@ -315,47 +631,58 @@ def export_results():
         'verdict': results.get('verdict', ''),
         'confidence_score': float(results.get('confidence', 0)),
         'ai_summary': results.get('summary', ''),
-        'analysis_time': results.get('analysis_time', 0)
+        'analysis_model': results.get('brain', ''),
+        'analysis_time': results.get('analysis_time', 0),
     }
     json_string = json.dumps(export_data, indent=2, default=str, ensure_ascii=False)
     st.download_button(
         label="📥 Download Analysis Report",
         data=json_string,
         file_name=f"credo_ai_analysis_{datetime.now().strftime('%Y%m%d_%H%M%S')}.json",
-        mime="application/json"
+        mime="application/json",
     )
     st.success("📄 Analysis report ready for download!")
 
 
 def render_analysis_results(results):
     st.markdown("### ✨ AI-Powered Analysis Summary")
-    st.markdown(f"""
-    <div class="summary-box">
-        {results['summary']}
-    </div>
-    """, unsafe_allow_html=True)
+    st.markdown(
+        f"""<div class="summary-box">{results['summary']}</div>""",
+        unsafe_allow_html=True,
+    )
     col1, col2 = st.columns(2, gap="large")
     with col1:
         st.markdown("### 🎯 Primary Verdict")
         verdict = results['verdict']
         confidence = results['confidence']
-        verdict_class = 'verdict-fake' if verdict == 'FAKE' else 'verdict-real'
-        st.markdown(f"""
-        <div class="verdict-container {verdict_class}">
-            <div class="verdict-text">{verdict}</div>
-        </div>
-        <div style="text-align: center; margin-top: 1rem; font-size: 1.5rem; font-weight: 600; color: #f1f5f9;">
-            {confidence:.1f}% Confidence
-        </div>
-        """, unsafe_allow_html=True)
+        verdict_class = 'verdict-real'
+        if verdict == 'FAKE':
+            verdict_class = 'verdict-fake'
+        elif verdict == 'UNCERTAIN':
+            verdict_class = 'verdict-uncertain'
+        st.markdown(
+            f"""
+            <div class="verdict-container {verdict_class}">
+                <div class="verdict-text">{verdict}</div>
+            </div>
+            <div style="text-align: center; margin-top: 1rem; font-size: 1.4rem; font-weight: 700; color: #334155;">
+                {confidence:.1f}% Confidence
+            </div>
+            """,
+            unsafe_allow_html=True,
+        )
+        st.progress(min(float(confidence) / 100.0, 1.0), text="Model confidence")
     with col2:
         st.markdown("### 📊 Analysis Details")
         st.metric("Processing Time", f"{results.get('analysis_time', 0):.2f}s")
         st.metric("Content Length", f"{len(results.get('input', '').split())} words")
-        st.metric("Analysis Method", "AI Analysis")
+        st.metric("Analysis Model", results.get('brain', 'Brain 2 (General)'))
+        if results.get('model_used'):
+            st.metric("Summary Layer", "Groq AI")
+        else:
+            st.metric("Summary Layer", "Built-in fallback")
 
 
-# Initialize session state
 if 'analysis_complete' not in st.session_state:
     st.session_state.analysis_complete = False
 if 'current_results' not in st.session_state:
@@ -363,43 +690,39 @@ if 'current_results' not in st.session_state:
 if 'analysis_history' not in st.session_state:
     st.session_state.analysis_history = []
 
-# API config for Gemini
-GOOGLE_API_KEY = os.getenv('GOOGLE_API_KEY')
-API_CONFIGURED = bool(GOOGLE_API_KEY and GENAI_AVAILABLE)
-if API_CONFIGURED:
-    try:
-        genai.configure(api_key=GOOGLE_API_KEY)
-    except Exception:
-        API_CONFIGURED = False
-
-# Sidebar and navigation
 with st.sidebar:
-    st.markdown("""
-    <div style="text-align: center; padding: 1rem 0; margin-bottom: 2rem;">
-        <h2 style="color: #6366f1; margin: 0;">🧠 Credo AI</h2>
-        <p style="color: #94a3b8; margin: 0.5rem 0 0 0; font-size: 0.9rem;">Truth Detection Platform</p>
-    </div>
-    """, unsafe_allow_html=True)
+    st.markdown(
+        """
+        <div style="text-align: center; padding: 1rem 0; margin-bottom: 2rem;">
+            <div style="font-size: 2.4rem; margin-bottom: 0.4rem;">🧠</div>
+            <h2 style="color: #4f46e5; margin: 0;">Credo AI</h2>
+            <p style="color: #94a3b8; margin: 0.5rem 0 0 0; font-size: 0.9rem;">Truth Detection Platform</p>
+        </div>
+        """,
+        unsafe_allow_html=True,
+    )
 
     page = st.radio(
         "Navigate:",
         ["🚀 Live Analysis", "📜 History", "ℹ️ About"],
-        key="navigation"
+        key="navigation",
     )
 
     if st.session_state.analysis_history:
         st.markdown("---")
         st.markdown("### 📈 Quick Stats")
         total = len(st.session_state.analysis_history)
-        fake_count = sum(1 for h in st.session_state.analysis_history if h.get('verdict') == 'FAKE')
+        fake_count = sum(
+            1 for h in st.session_state.analysis_history if h.get('verdict') == 'FAKE'
+        )
         st.metric("Total Analyses", total)
         if total > 0:
-            st.metric("Fake Rate", f"{(fake_count/total*100):.0f}%")
+            st.metric("Fake Rate", f"{(fake_count / total * 100):.0f}%")
 
     st.markdown("---")
     st.markdown("### 🔧 Status")
-    if API_CONFIGURED:
-        st.success("🟢 AI Enhanced")
+    if GROQ_AVAILABLE:
+        st.success("🟢 Groq Enhanced")
     else:
         st.warning("🟡 Basic Mode")
 
@@ -412,38 +735,45 @@ with st.sidebar:
         time.sleep(1)
         st.rerun()
 
-# Main app pages
 if page == "🚀 Live Analysis":
-    st.markdown("""
-    <div class="hero-container">
-        <h1 class="main-title">🧠 Credo AI Platform</h1>
-        <p class="hero-subtitle">
-            Next-generation misinformation detection powered by <strong>dual-AI architecture</strong>.
-            Analyze text, articles, and claims with unprecedented accuracy and insight.
-        </p>
-        <div class="metrics-container">
-            <div class="metric-card">
-                <span class="metric-value">99.9%</span>
-                <span class="metric-label">Accuracy</span>
-            </div>
-            <div class="metric-card">
-                <span class="metric-value">2</span>
-                <span class="metric-label">AI Brains</span>
-            </div>
-            <div class="metric-card">
-                <span class="metric-value">&lt;3s</span>
-                <span class="metric-label">Analysis Time</span>
+    st.markdown(
+        """
+        <div class="hero-container">
+            <h1 class="main-title">🧠 Credo AI Platform</h1>
+            <p class="hero-subtitle">
+                Next-generation misinformation detection powered by
+                <strong>dual-AI architecture</strong>. Analyze text, articles, and claims
+                with speed and insight.
+            </p>
+            <div class="metrics-container">
+                <div class="metric-card">
+                    <span class="metric-value">2</span>
+                    <span class="metric-label">AI Brains</span>
+                </div>
+                <div class="metric-card">
+                    <span class="metric-value">FAKE/REAL</span>
+                    <span class="metric-label">Verdict</span>
+                </div>
+                <div class="metric-card">
+                    <span class="metric-value">&lt;3s</span>
+                    <span class="metric-label">Analysis Time</span>
+                </div>
             </div>
         </div>
-    </div>
-    """, unsafe_allow_html=True)
+        """,
+        unsafe_allow_html=True,
+    )
 
-    if not API_CONFIGURED:
-        st.info("🔑 **Optional Setup:** Add GOOGLE_API_KEY in Space Settings → Variables and Secrets for enhanced AI summaries with Gemini. The platform works without it using intelligent fallback analysis.")
+    if not GROQ_AVAILABLE:
+        st.info(
+            "🔑 **Optional Setup:** Add GROQ_API_KEY in Space Settings → Variables and "
+            "Secrets for AI-powered summaries with Groq. The platform works without it "
+            "using intelligent fallback analysis."
+        )
 
     classifier_b1, classifier_b2 = load_ai_models()
     if classifier_b1 is None or classifier_b2 is None:
-        st.error("Failed to load AI models! Please try to restart the app or check logs.")
+        st.error("Failed to load AI models! Please restart the app or check logs.")
     else:
         render_analysis_interface(classifier_b1, classifier_b2)
 
@@ -456,8 +786,12 @@ elif page == "📜 History":
     st.markdown("# 📜 Analysis History")
     if st.session_state.analysis_history:
         total = len(st.session_state.analysis_history)
-        fake_count = sum(1 for h in st.session_state.analysis_history if h.get('verdict') == 'FAKE')
-        real_count = sum(1 for h in st.session_state.analysis_history if h.get('verdict') == 'REAL')
+        fake_count = sum(
+            1 for h in st.session_state.analysis_history if h.get('verdict') == 'FAKE'
+        )
+        real_count = sum(
+            1 for h in st.session_state.analysis_history if h.get('verdict') == 'REAL'
+        )
         st.markdown("### 📈 Summary Statistics")
         stat_cols = st.columns(3)
         with stat_cols[0]:
@@ -468,108 +802,129 @@ elif page == "📜 History":
             st.metric("Real Content", real_count)
         st.markdown("---")
         for i, result in enumerate(st.session_state.analysis_history):
-            with st.expander(f"#{i+1} - {result.get('verdict', 'Unknown')} | {result.get('input', 'No input')}", expanded=(i==0)):
+            with st.expander(
+                f"#{i + 1} - {result.get('verdict', 'Unknown')} | {result.get('input', 'No input')}",
+                expanded=(i == 0),
+            ):
                 render_analysis_results(result)
     else:
-        st.info("📚 **No Analysis History** - Your analysis history will appear here after you perform some fact-checking analyses. Start by going to the Live Analysis page and analyzing some content!")
+        st.info(
+            "📚 **No Analysis History** - Your analysis history will appear here after you "
+            "perform some fact-checking analyses. Start by going to the Live Analysis page "
+            "and analyzing some content!"
+        )
 
 elif page == "ℹ️ About":
     st.markdown("# 🔬 About Credo AI")
-    st.markdown("""
-    <div class="glass-card">
-        <h2 style="color: #6366f1; margin-bottom: 1rem;">🚀 Revolutionary Detection Technology</h2>
-        <p style="font-size: 1.2rem; color: #cbd5e1; line-height: 1.7;">
-            Credo AI represents a breakthrough in automated fact-checking, combining
-            <strong>two specialized neural networks</strong> with advanced language understanding
-            to deliver unparalleled accuracy in misinformation detection.
-        </p>
-    </div>
-    """, unsafe_allow_html=True)
+    st.markdown(
+        """
+        <div class="glass-card">
+            <h2 style="color: #4f46e5; margin-bottom: 1rem;">🚀 Revolutionary Detection Technology</h2>
+            <p style="font-size: 1.2rem; color: #475569; line-height: 1.7;">
+                Credo AI represents a breakthrough in automated fact-checking, combining
+                <strong>two specialized neural networks</strong> with advanced language
+                understanding to deliver rapid, transparent misinformation detection.
+            </p>
+        </div>
+        """,
+        unsafe_allow_html=True,
+    )
     tab1, tab2, tab3 = st.tabs(["🧠 AI Architecture", "📊 Performance", "🔬 Technology"])
 
     with tab1:
-        st.markdown("""
-        ### ⚡ Brain 2: The Specialist
-        - **Model:** `Arko007/fact-check1-v3-final`
-        - **Function:** Rapid FAKE/REAL binary classification
-        - **Training:** 80,000+ verified news articles
-        - **Performance:** 99.9% accuracy on benchmarks
-        - **Speed:** Sub-second inference time
-        
-        ### 🧠 Brain 1: The Nuance Expert
-        - **Model:** `Arko007/fake-news-liar-political`
-        - **Function:** Binary political fact-checking (US-centric)
-        - **Training:** LIAR dataset with focused binary labels
-        - **Performance:** ~71% accuracy
-        - **Specialization:** Short political statement classification
-        
-        ### ✨ Gemini Integration
-        - **Role:** Intelligent synthesis & explanation layer
-        - **Function:** Validates & optionally corrects classifications using real-time data
-        - **Value:** Enhances AI decisions invisibly to end users
-        """)
+        st.markdown(
+            """
+            ### ⚡ Brain 2: The Specialist
+            - **Model:** `Arko007/fact-check1-v3-final` (DeBERTa-v3-large)
+            - **Function:** Rapid FAKE/REAL binary classification
+            - **Training:** 50,000+ verified news articles + calibration passes
+            - **Speed:** Sub-second inference time
+
+            ### 🧠 Brain 1: The Political Expert
+            - **Model:** `Arko007/fake-news-liar-political` (RoBERTa-base)
+            - **Function:** Binary political fact-checking (US-centric)
+            - **Training:** LIAR dataset converted to binary
+            - **Performance:** ~71% accuracy
+            - **Specialization:** Short political statement classification
+
+            ### ⚡ Groq Integration
+            - **Role:** Intelligent synthesis & explanation layer
+            - **Model:** `qwen/qwen3.6-27b` (fallback: `openai/gpt-oss-120b`)
+            - **Function:** Validates classifications and explains verdicts in plain language
+            - **Speed:** Blazing-fast inference on Groq LPU hardware
+            """
+        )
 
     with tab2:
         st.markdown("### 📈 Performance Metrics")
         import pandas as pd
+
         metrics_data = {
             'Metric': ['Accuracy', 'Precision', 'Recall', 'F1-Score', 'Speed'],
             'Brain 1': ['71.4%', 'N/A', 'N/A', 'N/A', 'N/A'],
             'Brain 2': ['99.9%', '99.8%', '99.7%', '99.7%', '0.8s'],
-            'Combined': ['~95%', 'N/A', 'N/A', 'N/A', '<3s']
+            'Combined': ['~95%', 'N/A', 'N/A', 'N/A', '<3s'],
         }
-        st.dataframe(pd.DataFrame(metrics_data), use_container_width=True, hide_index=True)
+        st.dataframe(
+            pd.DataFrame(metrics_data),
+            width="stretch",
+            hide_index=True,
+        )
         st.success("🏆 Credo AI blends specialized models to maximize coverage and accuracy.")
 
     with tab3:
-        st.markdown("""
-        ### 🛠️ Technology Stack
-        
-        **🤖 Core AI/ML:**
-        - PyTorch deep learning framework
-        - Transformers library for model handling
-        - BERT-based and RoBERTa-based understanding
-        - Advanced fine-tuning techniques
-        
-        **🌐 Web & Integration:**
-        - Streamlit for responsive UI
-        - Beautiful Soup for web scraping
-        - Google Generative AI (Gemini 2.0)
-        - Tavily real-time information search
-        - Custom CSS for enhanced UX
-        
-        **⚡ Performance:**
-        - Intelligent caching system
-        - Memory-efficient processing
-        - Mobile-responsive design
-        - Privacy-first architecture
-        """)
+        st.markdown(
+            """
+            ### 🛠️ Technology Stack
+
+            **🤖 Core AI/ML:**
+            - PyTorch deep learning framework
+            - Hugging Face Transformers for model handling
+            - RoBERTa & DeBERTa-v3 fine-tuned classifiers
+
+            **🌐 Web & Integration:**
+            - Streamlit for responsive UI
+            - Beautiful Soup for web scraping
+            - Groq Cloud API (`qwen/qwen3.6-27b`)
+            - Tavily real-time information search
+            - Custom CSS for enhanced UX
+
+            **⚡ Performance:**
+            - Intelligent caching system
+            - Memory-efficient processing
+            - Mobile-responsive design
+            - Privacy-first architecture
+            """
+        )
 
-st.markdown("""
-<div class="footer-enhanced">
-    <div class="footer-features">
-        <div class="footer-feature">
-            <div class="footer-feature-icon">🏆</div>
-            <div class="footer-feature-text">Award Winning</div>
-        </div>
-        <div class="footer-feature">
-            <div class="footer-feature-icon">⚡</div>
-            <div class="footer-feature-text">Lightning Fast</div>
+st.markdown(
+    """
+    <div class="footer-enhanced">
+        <div class="footer-features">
+            <div class="footer-feature">
+                <div class="footer-feature-icon">🏆</div>
+                <div class="footer-feature-text">Award Winning</div>
+            </div>
+            <div class="footer-feature">
+                <div class="footer-feature-icon">⚡</div>
+                <div class="footer-feature-text">Lightning Fast</div>
+            </div>
+            <div class="footer-feature">
+                <div class="footer-feature-icon">🔒</div>
+                <div class="footer-feature-text">Privacy First</div>
+            </div>
+            <div class="footer-feature">
+                <div class="footer-feature-icon">🌍</div>
+                <div class="footer-feature-text">Global Impact</div>
+            </div>
         </div>
-        <div class="footer-feature">
-            <div class="footer-feature-icon">🔒</div>
-            <div class="footer-feature-text">Privacy First</div>
+        <div style="font-size: 0.9rem; opacity: 0.85;">
+            Built with ❤️ for Hack2Skill Hackathon 2025 | 🐉 Data Dragons Team
         </div>
-        <div class="footer-feature">
-            <div class="footer-feature-icon">🌍</div>
-            <div class="footer-feature-text">Global Impact</div>
+        <div style="font-size: 0.8rem; opacity: 0.6; margin-top: 0.5rem;">
+            Powered by Advanced AI • Making Truth Accessible to Everyone
         </div>
     </div>
-    <div style="font-size: 0.9rem; opacity: 0.8;">
-        Built with ❤️ for Hack2Skill Hackathon 2025 | 🐉 Data Dragons Team
-    </div>
-    <div style="font-size: 0.8rem; opacity: 0.6; margin-top: 0.5rem;">
-        Powered by Advanced AI • Making Truth Accessible to Everyone
-    </div>
-</div>
-""", unsafe_allow_html=True)
+    """,
+    unsafe_allow_html=True,
+)