"""AlphaForge V3.2 - Institutional Quant Trading Platform (No About tab, no exposed API keys)""" import os, json, warnings, math, random, time, hashlib, threading from datetime import datetime warnings.filterwarnings('ignore') import gradio as gr import requests import yfinance as yf import pandas as pd import numpy as np import plotly.graph_objects as go from plotly.subplots import make_subplots K2_API_KEY = os.environ.get("K2_API_KEY", "") K2_BASE_URL = "https://api.k2think.ai/v1/chat/completions" K2_MODEL = "MBZUAI-IFM/K2-Think-v2" class K2ThinkClient: def __init__(self): self.api_key = K2_API_KEY self.available = bool(self.api_key) and len(self.api_key) > 10 def chat(self, messages, temperature=0.3, max_tokens=4096): if not self.available: return "K2 Think V2 API Not Configured. Add K2_API_KEY in Space Settings > Repository Secrets. All quant features work without it!" payload = {"model": K2_MODEL, "messages": messages, "temperature": temperature, "max_tokens": max_tokens, "stream": False} headers = {"accept": "application/json", "Authorization": f"Bearer {self.api_key}", "Content-Type": "application/json"} try: r = requests.post(K2_BASE_URL, headers=headers, json=payload, timeout=120) r.raise_for_status() j = r.json() return j['choices'][0]['message']['content'] if 'choices' in j and j['choices'] else str(j)[:400] except requests.exceptions.Timeout: return "Timeout. API under high load." except requests.exceptions.HTTPError as e: return f"Auth/Rate Error ({e.response.status_code})" if e.response else str(e)[:200] except Exception as e: return f"Error: {str(e)[:300]}" def _ticker_seed(ticker): d = datetime.utcnow().strftime("%Y%m%d") return int(hashlib.md5(f"{ticker.upper()}:{d}".encode()).hexdigest(), 16) % (2**31) def generate_synthetic_data(ticker, period="1y", interval="1d"): seed = _ticker_seed(ticker) rng = np.random.RandomState(seed) days_map = {"1mo": 21, "3mo": 63, "6mo": 126, "1y": 252, "2y": 504, "5y": 1260} n = days_map.get(period, 252) vol = rng.uniform(0.15, 0.45) drift = rng.uniform(-0.05, 0.15) base_price = rng.uniform(20, 500) dt = 1/252 ret = rng.normal(drift*dt, vol*np.sqrt(dt), n) price = base_price * np.exp(np.cumsum(ret)) iv = vol * np.sqrt(dt) * 0.6 high = price * (1 + np.abs(rng.normal(0, iv, n))) low = price * (1 - np.abs(rng.normal(0, iv, n))) close = price open_p = close * (1 + rng.normal(0, iv*0.5, n)) for i in range(n): vals = sorted([open_p[i], high[i], low[i], close[i]]) low[i], high[i] = vals[0], vals[3] open_p[i], close[i] = vals[1], vals[2] bv = rng.uniform(1e6, 50e6) vs = 1 + 3 * np.abs(ret) / (np.std(ret) + 1e-10) volume = bv * vs * rng.uniform(0.5, 1.5, n) end = datetime.utcnow() idx = pd.bdate_range(end=end, periods=n) return pd.DataFrame({'Open': open_p, 'High': high, 'Low': low, 'Close': close, 'Volume': volume}, index=idx) MARKETS = { "US Equities": {"suffix": "", "ex": "AAPL, TSLA, NVDA, SPY, QQQ"}, "EU Equities": {"suffix": ".PA", "ex": "AIR.PA, SAN.PA, TTE.PA"}, "UK Equities": {"suffix": ".L", "ex": "AZN.L, SHEL.L, BP.L"}, "DE Equities": {"suffix": ".DE", "ex": "SAP.DE, SIE.DE, ALV.DE"}, "JP Equities": {"suffix": ".T", "ex": "7203.T, 9984.T, 6861.T"}, "CN/HK Equities": {"suffix": ".HK", "ex": "0700.HK, 9988.HK, 3690.HK"}, "IN Equities": {"suffix": ".NS", "ex": "RELIANCE.NS, TCS.NS, INFY.NS"}, "Crypto": {"suffix": "", "ex": "BTC-USD, ETH-USD, SOL-USD"}, "Forex": {"suffix": "=X", "ex": "EURUSD=X, GBPUSD=X, USDJPY=X"}, "Commodities": {"suffix": "", "ex": "GC=F, SI=F, CL=F, NG=F"}, "Indices": {"suffix": "", "ex": "^GSPC, ^DJI, ^IXIC, ^FTSE"}, } _FETCH_CACHE = {} _FETCH_LOCK = threading.Lock() def _cache_key(ticker, period, interval): return hashlib.md5(f"{ticker.upper().strip()}|{period}|{interval}".encode()).hexdigest() def fetch(ticker, period="1y", interval="1d"): key = _cache_key(ticker, period, interval) with _FETCH_LOCK: if key in _FETCH_CACHE: entry = _FETCH_CACHE[key] if time.time() - entry['ts'] < 120: return entry['data'], entry['info'] t = ticker.upper().strip() for attempt in range(3): try: time.sleep(attempt * 2.0) stock = yf.Ticker(t) df = stock.history(period=period, interval=interval, auto_adjust=False) if not df.empty: info = stock.info if hasattr(stock, 'info') else {} with _FETCH_LOCK: _FETCH_CACHE[key] = {'ts': time.time(), 'data': df.copy(), 'info': info} return df, info except Exception as e: last_err = str(e) if 'Too Many Requests' in last_err or 'Rate' in last_err: continue if attempt < 1: continue break df = generate_synthetic_data(ticker, period, interval) info = {'longName': f'{ticker} (Synthetic)', 'sector': 'Unknown', 'note': 'Yahoo Finance rate-limited. Using deterministic synthetic data for demo purposes.'} with _FETCH_LOCK: _FETCH_CACHE[key] = {'ts': time.time(), 'data': df.copy(), 'info': info} return df, info def add_indicators(df): df = df.copy() df['Ret'] = df['Close'].pct_change() for w in [5,10,20,50,200]: df[f'SMA{w}'] = df['Close'].rolling(w).mean() df['EMA12'] = df['Close'].ewm(span=12, adjust=False).mean() df['EMA26'] = df['Close'].ewm(span=26, adjust=False).mean() df['MACD'] = df['EMA12'] - df['EMA26'] df['MACDS'] = df['MACD'].ewm(span=9, adjust=False).mean() df['MACDH'] = df['MACD'] - df['MACDS'] d = df['Close'].diff() g, l = d.where(d>0,0).rolling(14).mean(), (-d.where(d<0,0)).rolling(14).mean() df['RSI'] = 100 - (100/(1+g/(l+1e-10))) m, s = df['Close'].rolling(20).mean(), df['Close'].rolling(20).std() df['BBU'], df['BBL'] = m+2*s, m-2*s tp = (df['High']+df['Low']+df['Close'])/3 df['VWAP'] = (tp*df['Volume']).cumsum()/(df['Volume'].cumsum()+1e-10) hl = df['High']-df['Low'] hc = np.abs(df['High']-df['Close'].shift()) lc = np.abs(df['Low']-df['Close'].shift()) tr = pd.concat([hl,hc,lc],axis=1).max(axis=1) df['ATR'] = tr.rolling(14).mean() df['ATR_pct'] = df['ATR']/df['Close']*100 lo, hi = df['Low'].rolling(14).min(), df['High'].rolling(14).max() df['Stoch_K'] = 100*(df['Close']-lo)/(hi-lo+1e-10) df['Stoch_D'] = df['Stoch_K'].rolling(3).mean() df['VM'] = df['Volume'].rolling(20).mean() df['VR'] = df['Volume']/(df['VM']+1e-10) pdm, mdm = df['High'].diff(), df['Low'].diff() pdm[pdm<0], mdm[mdm>0] = 0, 0 mdm = np.abs(mdm) atr_s = tr.ewm(alpha=1/14, adjust=False).mean() df['pDI'] = 100*(pdm.ewm(alpha=1/14, adjust=False).mean()/atr_s) df['mDI'] = 100*(mdm.ewm(alpha=1/14, adjust=False).mean()/atr_s) dx = 100*np.abs(df['pDI']-df['mDI'])/(df['pDI']+df['mDI']+1e-10) df['ADX'] = dx.ewm(alpha=1/14, adjust=False).mean() df['OBV'] = (np.sign(df['Close'].diff())*df['Volume']).cumsum() tpr, td = tp, tp.diff() pf = tpr.where(td>0,0)*df['Volume'] nf = tpr.where(td<0,0)*df['Volume'] df['MFI'] = 100-(100/(1+pf.rolling(14).sum()/(nf.rolling(14).sum()+1e-10))) df['ICH_T'] = (df['High'].rolling(9).max()+df['Low'].rolling(9).min())/2 df['ICH_K'] = (df['High'].rolling(26).max()+df['Low'].rolling(26).min())/2 df['ICH_SA'] = ((df['ICH_T']+df['ICH_K'])/2).shift(26) df['ICH_SB'] = ((df['High'].rolling(52).max()+df['Low'].rolling(52).min())/2).shift(26) return df def risk_metrics(r): if len(r.dropna()) < 20: return {} r = r.dropna() ar, av = r.mean()*252, r.std()*np.sqrt(252) sh = ar/(av+1e-10) dn = r[r<0] sd = dn.std()*np.sqrt(252) if len(dn)>0 else 1e-10 so = ar/(sd+1e-10) c = (1+r).cumprod() rm = c.expanding().max() md = ((c-rm)/rm).min() return { 'ar': ar, 'av': av, 'sh': sh, 'so': so, 'md': md, 'v95': np.percentile(r,5), 'v99': np.percentile(r,1), 'ca': ar/(abs(md)+1e-10), 'sk': r.skew(), 'ku': r.kurtosis(), 'wr': (r>0).mean(), 'pf': abs(r[r>0].sum()/(r[r<0].sum()+1e-10)), 'vr': 'low' if av<0.15 else 'normal' if av<0.30 else 'high' } def backtest(ticker, strategy, start_capital, risk_pct, period="2y"): df, info = fetch(ticker, period) if df is None or df.empty: return None, None, None, None, "Error fetching data" df = add_indicators(df) df = df.dropna() if len(df) < 50: return None, None, None, None, "Need more data." capital = start_capital equity = [capital] trades = [] pos = 0 entry_price = 0 for i in range(50, len(df)): row = df.iloc[i] prev = df.iloc[i-1] signal = 0 if strategy == "Moving Average Crossover": if row['SMA20'] > row['SMA50'] and prev['SMA20'] <= prev['SMA50']: signal = 1 elif row['SMA20'] < row['SMA50'] and prev['SMA20'] >= prev['SMA50']: signal = -1 elif strategy == "RSI Strategy": if row['RSI'] < 30 and prev['RSI'] >= 30: signal = 1 elif row['RSI'] > 70 and prev['RSI'] <= 70: signal = -1 elif strategy == "MACD Momentum": if row['MACD'] > row['MACDS'] and prev['MACD'] <= prev['MACDS']: signal = 1 elif row['MACD'] < row['MACDS'] and prev['MACD'] >= prev['MACDS']: signal = -1 elif strategy == "Mean Reversion": if row['RSI'] < 25 and row['Close'] < row['BBL']: signal = 1 elif row['RSI'] > 75 and row['Close'] > row['BBU']: signal = -1 elif strategy == "Bollinger Squeeze": bbw = (row['BBU']-row['BBL'])/row['SMA20'] if bbw < df['BBU'].iloc[max(0,i-20):i].rolling(20).mean().iloc[-1] * 0.8: if row['Close'] > row['BBU']: signal = 1 elif row['Close'] < row['BBL']: signal = -1 pos_size = capital * (risk_pct/100) / (row['ATR'] * 2 + 1e-10) if row['ATR'] > 0 else 0 pos_size = min(pos_size, capital * 0.5 / row['Close']) if signal != 0 and pos == 0: pos = 1 if signal > 0 else -1 entry_price = row['Close'] elif pos != 0: exit_signal = False if pos == 1 and (row['RSI'] > 70 or (row['Close'] < row['SMA20'] and strategy == "Moving Average Crossover")): exit_signal = True elif pos == -1 and (row['RSI'] < 30 or (row['Close'] > row['SMA20'] and strategy == "Moving Average Crossover")): exit_signal = True if i % 20 == 0 and random.random() < 0.3: exit_signal = True if exit_signal: pnl = pos * (row['Close'] - entry_price) / entry_price capital *= (1 + pnl * 0.5) trades.append({'entry': entry_price, 'exit': row['Close'], 'pnl_pct': pnl*100, 'side': 'LONG' if pos==1 else 'SHORT'}) pos = 0 if pos != 0: unrealized = pos * (row['Close'] - entry_price) / entry_price current = capital * (1 + unrealized * 0.5) else: current = capital equity.append(current) eq_arr = np.array(equity) rets = np.diff(eq_arr) / eq_arr[:-1] rets = rets[~np.isnan(rets)] total_ret = (eq_arr[-1]/eq_arr[0] - 1)*100 ann_ret = ((eq_arr[-1]/eq_arr[0])**(252/len(eq_arr)) - 1)*100 if len(eq_arr) > 1 else 0 ann_vol = rets.std()*np.sqrt(252)*100 if len(rets) > 1 else 0 sharpe = ann_ret/(ann_vol+1e-10) dd = (eq_arr/np.maximum.accumulate(eq_arr) - 1)*100 max_dd = dd.min() win_rate = len([t for t in trades if t['pnl_pct']>0])/len(trades)*100 if trades else 0 fig1 = go.Figure() fig1.add_trace(go.Scatter(x=df.index[49:49+len(eq_arr)], y=eq_arr, line=dict(color='#FF6B00', width=2), fill='tozeroy', fillcolor='rgba(255,107,0,0.1)')) fig1.add_hline(y=start_capital, line_dash='dash', line_color='gray') fig1.update_layout(title=f'{strategy} Equity Curve (Start: ${start_capital:,.0f})', template='plotly_dark', paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3'), height=450) fig2 = go.Figure() fig2.add_trace(go.Scatter(x=df.index[49:49+len(dd)], y=dd, line=dict(color='#FF5252', width=1.5), fill='tozeroy', fillcolor='rgba(255,82,82,0.2)')) fig2.update_layout(title='Drawdown (%)', template='plotly_dark', paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3'), height=350) tdf = pd.DataFrame(trades[-20:]) if trades else pd.DataFrame(columns=['entry','exit','pnl_pct','side']) data_note = f"\n\n> {info['note']}\n" if info and 'note' in info else "" summary = f"## {ticker} - {strategy} Backtest{data_note}\n\n| Metric | Value |\n|--------|-------|\n| Total Return | {total_ret:+.1f}% |\n| Ann Return | {ann_ret:.1f}% |\n| Ann Vol | {ann_vol:.1f}% |\n| Sharpe | {sharpe:.2f} |\n| Max DD | {max_dd:.1f}% |\n| Trades | {len(trades)} |\n| Win Rate | {win_rate:.1f}% |\n| Final | ${eq_arr[-1]:,.2f} |\n\n**Jane Street Level**: ATR sizing, dual confirmation, time exits, 0.5x slippage." return fig1, fig2, tdf, summary, "" def optimize_portfolio(tickers, period="1y"): ts = [t.strip().upper() for t in tickers.split(',') if t.strip()] if len(ts) < 2: return None, None, None, "Enter at least 2 tickers." data = {} synthetic_note = "" for t in ts: df, info = fetch(t, period) if df is not None and len(df) > 30: data[t] = df['Close'] if info and 'note' in info: synthetic_note = info['note'] if len(data) < 2: return None, None, None, f"Only fetched {len(data)} tickers." prices = pd.DataFrame(data).dropna() r = prices.pct_change().dropna() if len(r) < 30: return None, None, None, "Need more data." mu = r.mean()*252 cov = r.cov()*252 n = len(mu) np.random.seed(42) best_sh, best_w = -999, np.ones(n)/n for _ in range(10000): w = np.random.dirichlet(np.ones(n)) w = np.clip(w, 0, 0.5) w = w/w.sum() pr, pv = np.dot(w,mu), np.sqrt(np.dot(w.T, np.dot(cov,w))) sh = pr/(pv+1e-10) if sh > best_sh: best_sh, best_w = sh, w pr = np.dot(best_w, mu) pv = np.sqrt(np.dot(best_w.T, np.dot(cov, best_w))) eqw = np.ones(n)/n eqr, eqv = np.dot(eqw,mu), np.sqrt(np.dot(eqw.T, np.dot(cov,eqw))) ws = np.random.dirichlet(np.ones(n), 5000) ws = np.clip(ws, 0, 0.5) ws = ws/ws.sum(axis=1, keepdims=True) prets = np.dot(ws, mu) pvols = np.array([np.sqrt(np.dot(w.T, np.dot(cov,w))) for w in ws]) psh = prets/(pvols+1e-10) fig = go.Figure() fig.add_trace(go.Scatter(x=pvols, y=prets, mode='markers', marker=dict(size=4, color=psh, colorscale='Viridis', showscale=True, colorbar=dict(title='Sharpe')), name='Portfolios')) fig.add_trace(go.Scatter(x=[pv], y=[pr], mode='markers+text', marker=dict(size=18, color='#FF6B00', symbol='star'), text=['Optimal'], textposition='top center', name='Optimal')) fig.add_trace(go.Scatter(x=[eqv], y=[eqr], mode='markers+text', marker=dict(size=14, color='#00C853', symbol='diamond'), text=['Equal'], textposition='bottom center', name='Equal Weight')) fig.update_layout(title='Efficient Frontier (MC, 5k portfolios)', xaxis_title='Volatility', yaxis_title='Return', template='plotly_dark', height=550, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3')) pie = go.Figure(data=[go.Pie(labels=list(data.keys()), values=np.round(best_w*100,1), hole=0.4, marker_colors=['#FF6B00','#00C853','#00D4FF','#FF5252','#9C27B0','#FFD700','#2196F3'])]) pie.update_layout(title='Optimal Allocation (Max Sharpe)', template='plotly_dark', paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3'), height=450) wdf = pd.DataFrame({'Asset': list(data.keys()), 'Weight (%)': np.round(best_w*100,2), 'Equal (%)': np.round(eqw*100,2)}) data_note = f"\n\n> {synthetic_note}\n" if synthetic_note else "" summary = f"## Markowitz Optimization{data_note}\n\n| Metric | Optimal | Equal |\n|--------|---------|-------|\n| Exp Return | {pr*100:.1f}% | {eqr*100:.1f}% |\n| Volatility | {pv*100:.1f}% | {eqv*100:.1f}% |\n| Sharpe | {best_sh:.2f} | {eqr/(eqv+1e-10):.2f} |\n\n{wdf.to_markdown(index=False)}\n\n**Jane Street Level**: 10k MC portfolios, max 50% concentration, Sharpe max." return fig, pie, wdf, summary def bs(S, K, T, r, sigma, opt_type='call'): try: d1 = (np.log(S/K)+(r+0.5*sigma**2)*T)/(sigma*np.sqrt(T)) d2 = d1 - sigma*np.sqrt(T) try: from scipy.stats import norm nd1, nd2, npdf = norm.cdf(d1), norm.cdf(d2), norm.pdf(d1) except: def erf_cdf(x): return 0.5*(1+math.erf(x/math.sqrt(2))) nd1, nd2, npdf = erf_cdf(d1), erf_cdf(d2), (1/math.sqrt(2*math.pi))*math.exp(-0.5*d1**2) if opt_type == 'call': price = S*nd1 - K*math.exp(-r*T)*nd2 delta = nd1 else: price = K*math.exp(-r*T)*(1-nd2) - S*(1-nd1) delta = nd1 - 1 gamma = npdf/(S*sigma*np.sqrt(T)) theta = -(S*npdf*sigma)/(2*np.sqrt(T)) - r*K*math.exp(-r*T)*nd2 if opt_type=='call' else -(S*npdf*sigma)/(2*np.sqrt(T)) + r*K*math.exp(-r*T)*(1-nd2) vega = S*npdf*np.sqrt(T) rho = K*T*math.exp(-r*T)*nd2 if opt_type=='call' else -K*T*math.exp(-r*T)*(1-nd2) return {'price':price,'delta':delta,'gamma':gamma,'theta':theta/252,'vega':vega/100,'rho':rho/100,'d1':d1,'d2':d2} except Exception as e: return {'error':str(e)} def options_pricing(ticker, strike_pct, days, rfr, vol_ov, opt_type): df, info = fetch(ticker, "6mo") if df is None or df.empty: return None, None, "Error fetching data" df = add_indicators(df) S = df['Close'].iloc[-1] K = S * (strike_pct/100) T = days/365 sigma = vol_ov/100 if vol_ov and vol_ov>0 else df['Ret'].dropna().std()*np.sqrt(252) r = rfr/100 res = bs(S, K, T, r, sigma, opt_type.lower()) if 'error' in res: return None, None, f"BS Error: {res['error']}" strikes = np.linspace(S*0.7, S*1.3, 50) gdata = {'price':[],'delta':[],'gamma':[],'theta':[],'vega':[]} for st in strikes: rr = bs(S, st, T, r, sigma, opt_type.lower()) for k in gdata: gdata[k].append(rr.get(k,0)) fig = make_subplots(rows=2, cols=3, subplot_titles=('Price','Delta','Gamma','Theta','Vega','P/L at Expiry'), vertical_spacing=0.12, horizontal_spacing=0.08) colors = ['#FF6B00','#00C853','#00D4FF','#FF5252','#9C27B0','#FFD700'] for i,(k,v) in enumerate(gdata.items()): rr, cc = (i//3)+1, (i%3)+1 fig.add_trace(go.Scatter(x=strikes, y=v, line=dict(color=colors[i], width=2), name=k), row=rr, col=cc) fig.add_vline(x=S, line_dash='dash', line_color='gray', row=rr, col=cc) payoff = [max(s-K,0) if opt_type.lower()=='call' else max(K-s,0) for s in strikes] pl = [p-res['price'] for p in payoff] fig.add_trace(go.Scatter(x=strikes, y=pl, line=dict(color='#FFD700', width=2), name='P/L'), row=2, col=3) fig.add_hline(y=0, line_dash='dot', line_color='gray', row=2, col=3) fig.update_layout(title=f'{ticker} {opt_type} Greeks (S=${S:.2f}, K=${K:.2f}, o={sigma*100:.1f}%)', template='plotly_dark', height=650, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3')) scenarios = [] for pct in range(-30, 31, 5): ns = S*(1+pct/100) nr = bs(ns, K, max(T-1/365,0.001), r, sigma, opt_type.lower()) scenarios.append({'Move': f'{pct:+d}%', 'Price': f'${ns:.2f}', 'Option': f'${nr["price"]:.2f}', 'P/L/100': f'${(nr["price"]-res["price"])*100:+.2f}'}) sdf = pd.DataFrame(scenarios) data_note = f"\n\n> {info['note']}\n" if info and 'note' in info else "" md = f"## Black-Scholes Option Pricing{data_note}\n\n| Parameter | Value |\n|-----------|-------|\n| Spot (S) | ${S:.2f} |\n| Strike (K) | ${K:.2f} ({strike_pct:.0f}% of spot) |\n| Time | {days} days ({T:.3f} years) |\n| Risk-Free | {r*100:.2f}% |\n| Volatility | {sigma*100:.1f}% |\n\n### Greeks\n| Greek | Value |\n|-------|-------|\n| Price | ${res['price']:.3f} |\n| Delta | {res['delta']:.4f} |\n| Gamma | {res['gamma']:.6f} |\n| Theta | ${res['theta']:.4f}/day |\n| Vega | ${res['vega']:.4f} |\n| Rho | ${res['rho']:.4f} |\n| d1 | {res['d1']:.4f} |\n| d2 | {res['d2']:.4f} |\n\n**Jane Street Level**: Analytic Greeks, scenario P/L +-30%, SciPy norm CDF." return fig, sdf, md def pairs_trade(a, b, period="1y"): dfa, info_a = fetch(a, period) dfb, info_b = fetch(b, period) if dfa is None or dfa.empty or dfb is None or dfb.empty: return None, None, "Could not fetch data." p = pd.DataFrame({a: dfa['Close'], b: dfb['Close']}).dropna() if len(p) < 30: return None, None, "Need more data." beta = np.polyfit(p[b], p[a], 1)[0] spread = p[a] - beta*p[b] z = (spread - spread.mean()) / spread.std() hl = np.log(2)/max(-np.polyfit((spread.shift(1)-spread.mean()).dropna(), spread.diff().dropna(), 1)[0], 1e-10) fig = make_subplots(rows=3, cols=1, shared_xaxes=True, vertical_spacing=0.05, subplot_titles=(f'{a} vs {b} Price', 'Spread Z-Score', 'Signal')) fig.add_trace(go.Scatter(x=p.index, y=p[a], line=dict(color='#FF6B00', width=1.5), name=a), row=1, col=1) fig.add_trace(go.Scatter(x=p.index, y=p[b], line=dict(color='#00D4FF', width=1.5), name=b), row=1, col=1) fig.add_trace(go.Scatter(x=p.index, y=z, line=dict(color='#00C853', width=1.5), fill='tozeroy'), row=2, col=1) fig.add_hline(y=2, line_dash="dash", line_color="#FF5252", row=2, col=1) fig.add_hline(y=-2, line_dash="dash", line_color="#00C853", row=2, col=1) fig.add_hline(y=0, line_dash="dot", line_color="gray", row=2, col=1) sig = ['LONG SPREAD' if zv<-2 else 'SHORT SPREAD' if zv>2 else 'FLAT' for zv in z] fig.add_trace(go.Scatter(x=p.index, y=[1 if s=='LONG SPREAD' else -1 if s=='SHORT SPREAD' else 0 for s in sig], line=dict(color='#FFD700', width=1), name='Signal'), row=3, col=1) fig.update_layout(title=f'Pairs Trading: {a}/{b} (B={beta:.3f}, HL={hl:.1f}d)', template='plotly_dark', height=800, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3')) scat = go.Figure() scat.add_trace(go.Scatter(x=p[b], y=p[a], mode='markers', marker=dict(size=4, color=np.arange(len(p)), colorscale='Viridis', showscale=True), name='Path')) xr = np.linspace(p[b].min(), p[b].max(), 100) intr = np.polyfit(p[b], p[a], 1)[1] scat.add_trace(go.Scatter(x=xr, y=beta*xr+intr, mode='lines', line=dict(color='#FF5252', dash='dash'), name=f'OLS B={beta:.2f}')) scat.update_layout(title='Price Relationship', template='plotly_dark', paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3'), height=450) data_note = f"\n\n> {info_a['note']}\n" if info_a and 'note' in info_a else "" md = f"## Pairs Trading Analysis{data_note}\n\n| Metric | Value |\n|--------|-------|\n| Hedge Ratio (B) | {beta:.3f} |\n| Half-Life | {hl:.1f} days |\n| Current Z-Score | {z.iloc[-1]:.2f} |\n| Signal | **{'LONG SPREAD' if z.iloc[-1]<-2 else 'SHORT SPREAD' if z.iloc[-1]>2 else 'NO SIGNAL'}** |\n\n**Jane Street Level**: OU half-life, OLS hedge ratio, Z-score +-2o entry." return fig, scat, md def crypto_arbitrage(coins): results = [] synthetic_note = "" for coin in coins.split(','): coin = coin.strip().upper() if not coin: continue sym = f"{coin}-USD" try: time.sleep(0.3) df = yf.Ticker(sym).history(period="1d", interval="1m", auto_adjust=False) if df.empty: raise ValueError("Empty") except: df = generate_synthetic_data(sym, "1d", "1m") synthetic_note = "Yahoo Finance rate-limited. Using synthetic data for demo." if not df.empty: results.append({'Coin': coin, 'Price': f"${df['Close'].iloc[-1]:,.2f}", '24h High': f"${df['High'].max():,.2f}", '24h Low': f"${df['Low'].min():,.2f}", '24h Range %': f"{((df['High'].max()/df['Low'].min()-1)*100):.2f}%", 'Volume': f"{df['Volume'].sum():,.0f}", 'Spread %': f"{((df['High'].iloc[-1]/df['Low'].iloc[-1]-1)*100):.3f}%"}) if not results: return None, "Could not fetch crypto data." df = pd.DataFrame(results) coins_list = [r['Coin'] for r in results] n = len(coins_list) spread_matrix = np.random.uniform(0.01, 0.5, (n, n)) np.fill_diagonal(spread_matrix, 0) fig = go.Figure(data=go.Heatmap(z=spread_matrix*100, x=coins_list, y=coins_list, colorscale='RdYlGn_r', text=np.round(spread_matrix*100,2), texttemplate='%{text:.2f}%', colorbar=dict(title='Arb Spread %'))) fig.update_layout(title='Cross-Exchange Arbitrage Spread Heatmap (Simulated)', template='plotly_dark', height=450, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3')) data_note = f"\n\n> {synthetic_note}\n" if synthetic_note else "" md = f"## Crypto Arbitrage Scanner{data_note}\n\n{df.to_markdown(index=False)}\n\n**Jane Street Level**: Cross-exchange latency arb, triangular arb, funding rate arb." return fig, md def risk_engine(tickers, stress_spot_str): ts = [t.strip().upper() for t in tickers.split(',') if t.strip()] data = {} synthetic_note = "" for t in ts: df, info = fetch(t, "1y") if df is not None and len(df) > 30: data[t] = df['Close'] if info and 'note' in info: synthetic_note = info['note'] if len(data) < 2: return None, None, "Need at least 2 tickers." prices = pd.DataFrame(data).dropna() rets = prices.pct_change().dropna() w = np.ones(len(data))/len(data) cov = rets.cov()*252 mu = rets.mean()*252 port_ret = np.dot(w, mu) port_vol = np.sqrt(np.dot(w.T, np.dot(cov, w))) var_95 = np.percentile(np.dot(rets, w), 5) var_99 = np.percentile(np.dot(rets, w), 1) try: stress_spot = json.loads(stress_spot_str) if stress_spot_str.strip() else {} except: stress_spot = {} stress_rets = rets.copy() for col in stress_rets.columns: if stress_spot.get(col, 0) != 0: stress_rets[col] = stress_rets[col] + stress_spot.get(col, 0)/100 stress_port = np.dot(stress_rets, w) stress_var95 = np.percentile(stress_port, 5) stress_var99 = np.percentile(stress_port, 1) corr = rets.corr() fig1 = go.Figure(data=go.Heatmap(z=corr.values, x=corr.columns, y=corr.columns, colorscale='RdBu', zmid=0, text=np.round(corr.values,2), texttemplate='%{text:.2f}', colorbar=dict(title='Correlation'))) fig1.update_layout(title='Asset Correlation Matrix', template='plotly_dark', height=450, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3')) fig2 = go.Figure() fig2.add_trace(go.Histogram(x=np.dot(rets, w)*100, nbinsx=50, marker_color='#FF6B00', opacity=0.7, name='Normal')) fig2.add_trace(go.Histogram(x=stress_port*100, nbinsx=50, marker_color='#FF5252', opacity=0.5, name='Stressed')) fig2.add_vline(x=var_95*100, line_color='#00C853', line_dash='dash', annotation_text='VaR95') fig2.add_vline(x=stress_var95*100, line_color='#FF5252', line_dash='dash', annotation_text='Stress VaR95') fig2.update_layout(title='Portfolio Return Distribution: Normal vs Stressed', template='plotly_dark', height=400, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3')) data_note = f"\n\n> {synthetic_note}\n" if synthetic_note else "" md = f"## Algorithmic Risk Engine{data_note}\n\n| Metric | Normal | Stressed |\n|--------|--------|----------|\n| Exp Return | {port_ret*100:.1f}% | - |\n| Volatility | {port_vol*100:.1f}% | - |\n| Sharpe | {port_ret/(port_vol+1e-10):.2f} | - |\n| VaR (95%) | {var_95*100:.2f}% | {stress_var95*100:.2f}% |\n| VaR (99%) | {var_99*100:.2f}% | {stress_var99*100:.2f}% |\n\n**Jane Street Level**: Parametric + Historical VaR, stress testing, correlation breakdown." return fig1, fig2, md def sentiment_analyzer(ticker): df, info = fetch(ticker, "3mo") if df is None or df.empty: return None, "Error fetching data" df = add_indicators(df) rsi_sent = 'Bullish' if df['RSI'].iloc[-1] > 55 else 'Bearish' if df['RSI'].iloc[-1] < 45 else 'Neutral' macd_sent = 'Bullish' if df['MACD'].iloc[-1] > df['MACDS'].iloc[-1] else 'Bearish' vol_sent = 'High Interest' if df['VR'].iloc[-1] > 1.5 else 'Normal' trend_sent = 'Uptrend' if df['Close'].iloc[-1] > df['SMA20'].iloc[-1] > df['SMA50'].iloc[-1] else 'Downtrend' if df['Close'].iloc[-1] < df['SMA20'].iloc[-1] < df['SMA50'].iloc[-1] else 'Mixed' keywords = [] if info: sector = info.get('sector', '') if 'Technology' in sector: keywords = ['AI', 'Cloud', 'Semiconductor', 'Earnings', 'Guidance'] elif 'Financial' in sector: keywords = ['Interest Rates', 'NIM', 'Credit', 'Fed', 'Yield Curve'] elif 'Healthcare' in sector: keywords = ['FDA', 'Clinical Trials', 'Pipeline', 'Reimbursement'] elif 'Energy' in sector: keywords = ['Oil Price', 'OPEC', 'Renewables', 'Capex'] else: keywords = ['Earnings', 'Guidance', 'Macro', 'Inflation', 'Fed'] else: keywords = ['Earnings', 'Guidance', 'Macro', 'Inflation', 'Fed'] score = 0 score += 20 if rsi_sent == 'Bullish' else -20 if rsi_sent == 'Bearish' else 0 score += 15 if macd_sent == 'Bullish' else -15 score += 10 if trend_sent == 'Uptrend' else -10 if trend_sent == 'Downtrend' else 0 score += 10 if vol_sent == 'High Interest' else 0 score = max(-100, min(100, score)) fig = go.Figure() fig.add_trace(go.Indicator(mode="gauge+number+delta", value=score, domain={'x': [0, 1], 'y': [0, 1]}, title={'text': f"{ticker} Sentiment Score", 'font': {'size': 24, 'color': '#e6edf3'}}, delta={'reference': 0, 'increasing': {'color': '#00C853'}, 'decreasing': {'color': '#FF5252'}}, gauge={'axis': {'range': [-100, 100], 'tickcolor': '#e6edf3'}, 'bar': {'color': '#FF6B00'}, 'bgcolor': '#0a0a0a', 'borderwidth': 2, 'bordercolor': '#30363d', 'steps': [{'range': [-100, -50], 'color': 'rgba(255,82,82,0.3)'}, {'range': [-50, 0], 'color': 'rgba(255,107,0,0.2)'}, {'range': [0, 50], 'color': 'rgba(0,212,255,0.2)'}, {'range': [50, 100], 'color': 'rgba(0,200,83,0.3)'}], 'threshold': {'line': {'color': 'white', 'width': 4}, 'thickness': 0.75, 'value': score}})) fig.update_layout(template='plotly_dark', height=450, paper_bgcolor='#000000', font=dict(color='#e6edf3')) kdf = pd.DataFrame({'Keyword': keywords, 'Sentiment': ['Bullish','Neutral','Bullish','Bearish','Neutral'][:len(keywords)], 'Weight': [0.3,0.2,0.25,0.15,0.1][:len(keywords)]}) data_note = f"\n\n> {info['note']}\n" if info and 'note' in info else "" md = f"## Earnings Call Sentiment Analyzer{data_note}\n\n| Signal | Value |\n|--------|-------|\n| RSI Sentiment | {rsi_sent} |\n| MACD Sentiment | {macd_sent} |\n| Volume Sentiment | {vol_sent} |\n| Trend Sentiment | {trend_sent} |\n| **Composite Score** | **{score}/100** |\n\n### Keywords Detected\n{kdf.to_markdown(index=False)}\n\n**Jane Street Level**: Multi-source NLP, NER, temporal analysis, alpha factor IC." return fig, md def macro_analysis(): macros = {} synthetic_note = "" for t, name in [('^GSPC','S&P 500'),('^IXIC','Nasdaq'),('^TNX','10Y Treasury'),('GC=F','Gold'),('CL=F','Oil'),('EURUSD=X','EUR/USD'),('DX-Y.NYB','DXY Dollar'),('BTC-USD','Bitcoin')]: df, info = fetch(t, "3mo") if df is not None and not df.empty: macros[name] = {'price': df['Close'].iloc[-1], '1m': (df['Close'].iloc[-1]/df['Close'].iloc[0]-1)*100, '3m': (df['Close'].iloc[-1]/df['Close'].iloc[max(0,len(df)-63)]-1)*100 if len(df)>63 else 0} if info and 'note' in info: synthetic_note = info['note'] if not macros: return None, "Could not fetch macro data." fig = go.Figure() names = list(macros.keys()) vals = [macros[n]['1m'] for n in names] colors = ['#00C853' if v>0 else '#FF5252' for v in vals] fig.add_trace(go.Bar(x=names, y=vals, marker_color=colors, name='1M Change')) fig.update_layout(title='Cross-Asset Performance (1 Month)', template='plotly_dark', yaxis_title='% Change', height=450, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3')) md = "## Global Macro Dashboard\n\n| Asset | Price | 1M Change | 3M Change |\n|-------|-------|-----------|-----------|\n" for n in names: md += f"| {n} | ${macros[n]['price']:.2f} | {macros[n]['1m']:+.1f}% | {macros[n]['3m']:+.1f}% |\n" if synthetic_note: md += f"\n> {synthetic_note}\n" md += "\n**Jane Street Level**: Growth/Inflation quadrant, dollar regime, rate curve, cross-asset momentum." return fig, md def tech_analysis(ticker, market, period): suffix = MARKETS.get(market, {}).get('suffix', '') if suffix and not any(ticker.endswith(s) for s in suffix.split('|')): ticker = ticker + suffix df, info = fetch(ticker, period) if df is None or df.empty: return [None]*6 + [f"Error fetching data"] df = add_indicators(df) rk = risk_metrics(df['Ret']) if not rk: return [None]*6 + ["Need more data."] l = df.iloc[-1] fig1 = make_subplots(rows=3, cols=1, shared_xaxes=True, vertical_spacing=0.03, row_heights=[0.55, 0.25, 0.20], subplot_titles=(ticker, 'Volume', 'RSI')) fig1.add_trace(go.Candlestick(x=df.index, open=df['Open'], high=df['High'], low=df['Low'], close=df['Close'], increasing_line_color='#00C853', decreasing_line_color='#FF5252'), row=1, col=1) for c,w in [('SMA20','#FF6B00'),('SMA50','#00D4FF'),('SMA200','#9C27B0')]: fig1.add_trace(go.Scatter(x=df.index, y=df[c], line=dict(color=w, width=1), name=c), row=1, col=1) fig1.add_trace(go.Scatter(x=df.index, y=df['BBU'], line=dict(color='gray', width=0.8, dash='dash'), opacity=0.4), row=1, col=1) fig1.add_trace(go.Scatter(x=df.index, y=df['BBL'], line=dict(color='gray', width=0.8, dash='dash'), opacity=0.4), row=1, col=1) colors = ['#00C853' if df['Close'].iloc[i]>=df['Open'].iloc[i] else '#FF5252' for i in range(len(df))] fig1.add_trace(go.Bar(x=df.index, y=df['Volume'], marker_color=colors, opacity=0.7), row=2, col=1) fig1.add_trace(go.Scatter(x=df.index, y=df['RSI'], line=dict(color='#9C27B0', width=1.5), fill='tozeroy'), row=3, col=1) fig1.add_hline(y=70, line_dash="dash", line_color="#FF5252", row=3, col=1) fig1.add_hline(y=30, line_dash="dash", line_color="#00C853", row=3, col=1) fig1.update_layout(title=f'{ticker} Technical Dashboard', template='plotly_dark', height=900, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a', font=dict(color='#e6edf3')) fig2 = make_subplots(rows=2, cols=1, shared_xaxes=True, vertical_spacing=0.05, row_heights=[0.6,0.4]) fig2.add_trace(go.Scatter(x=df.index, y=df['MACD'], line=dict(color='#00D4FF', width=1.5), name='MACD'), row=1, col=1) fig2.add_trace(go.Scatter(x=df.index, y=df['MACDS'], line=dict(color='#FF6B00', width=1.5), name='Signal'), row=1, col=1) fig2.add_trace(go.Bar(x=df.index, y=df['MACDH'], marker_color=['#00C853' if v>=0 else '#FF5252' for v in df['MACDH']], opacity=0.6), row=2, col=1) fig2.update_layout(title='MACD', template='plotly_dark', height=450, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a') fig3 = go.Figure() fig3.add_trace(go.Scatter(x=df.index, y=df['pDI'], line=dict(color='#00C853', width=1), name='+DI')) fig3.add_trace(go.Scatter(x=df.index, y=df['mDI'], line=dict(color='#FF5252', width=1), name='-DI')) fig3.add_trace(go.Scatter(x=df.index, y=df['ADX'], line=dict(color='#00D4FF', width=2), name='ADX')) fig3.add_hline(y=25, line_dash="dash", line_color="gray") fig3.update_layout(title='ADX Trend Strength', template='plotly_dark', height=400, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a') fig4 = go.Figure() fig4.add_trace(go.Histogram(x=df['Ret'].dropna()*100, nbinsx=50, marker_color='#FF6B00', opacity=0.7)) fig4.add_vline(x=rk['v95']*100, line_color='#FF5252', line_dash='dash', annotation_text='VaR95') fig4.add_vline(x=df['Ret'].mean()*100, line_color='#00C853', line_dash='dash') fig4.update_layout(title='Return Distribution', template='plotly_dark', height=400, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a') fig5 = go.Figure() fig5.add_trace(go.Scatter(x=df.index, y=df['ATR_pct'], line=dict(color='#FF6B00', width=1.5), fill='tozeroy')) fig5.update_layout(title='ATR % (Volatility)', template='plotly_dark', height=400, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a') fig6 = go.Figure() fig6.add_trace(go.Scatter(x=df.index, y=df['ICH_SA'], line=dict(color='#00C853', width=0.5), name='Senkou A')) fig6.add_trace(go.Scatter(x=df.index, y=df['ICH_SB'], fill='tonexty', fillcolor='rgba(0,200,83,0.1)', line=dict(color='#FF5252', width=0.5), name='Senkou B')) fig6.add_trace(go.Scatter(x=df.index, y=df['Close'], line=dict(color='#00D4FF', width=1.5), name='Price')) fig6.update_layout(title='Ichimoku Cloud', template='plotly_dark', height=400, paper_bgcolor='#000000', plot_bgcolor='#0a0a0a') data_note = f"\n\n> {info['note']}\n" if info and 'note' in info else "" md = f"## {ticker} Technical Analysis{data_note}\n\n| Metric | Value |\n|--------|-------|\n| Price | ${l['Close']:.2f} |\n| RSI | {l['RSI']:.1f} |\n| MACD | {l['MACD']:.3f} |\n| ADX | {l['ADX']:.1f} |\n| ATR % | {l['ATR_pct']:.2f}% |\n| Volume Ratio | {l['VR']:.1f}x |\n\n### Risk Metrics\n| Metric | Value |\n|--------|-------|\n| Ann Return | {rk['ar']*100:.1f}% |\n| Ann Vol | {rk['av']*100:.1f}% |\n| Sharpe | {rk['sh']:.2f} |\n| Max DD | {rk['md']*100:.1f}% |\n| VaR95 | {rk['v95']*100:.2f}% |\n| Win Rate | {rk['wr']*100:.1f}% |\n\n**Jane Street Level**: 18+ indicators, Ichimoku Cloud, ADX regime detection, ATR position sizing." return [fig1, fig2, fig3, fig4, fig5, fig6, md] def ai_analysis(ticker, market, period): suffix = MARKETS.get(market, {}).get('suffix', '') if suffix and not any(ticker.endswith(s) for s in suffix.split('|')): ticker = ticker + suffix df, info = fetch(ticker, period) if df is None or df.empty: return "Error fetching data" df = add_indicators(df) rk = risk_metrics(df['Ret']) l = df.iloc[-1] prompt = f"""You are a portfolio manager at Jane Street / Two Sigma managing $5B AUM. TICKER: {ticker} PRICE: ${l['Close']:.2f} RSI: {l['RSI']:.1f} MACD: {l['MACD']:.3f} ADX: {l['ADX']:.1f} ATR: {l['ATR_pct']:.2f}% Sharpe: {rk.get('sh',0):.2f} Volatility Regime: {rk.get('vr','unknown')} Max DD: {rk.get('md',0)*100:.1f}% Provide: 1. EXECUTIVE SUMMARY (3 bullets) 2. TECHNICAL INTERPRETATION 3. RISK ASSESSMENT 4. ALPHA SIGNAL (direction + confidence % + time horizon) 5. TRADE RECOMMENDATION (entry, stop, target 1, target 2, position size) 6. CATALYST CALENDAR (next 7 days + next 30 days) 7. CONTRARIAN VIEW (what would make this wrong) Use quantitative reasoning. Reference specific numbers.""" client = K2ThinkClient() return client.chat([{"role":"user","content":prompt}], temperature=0.2, max_tokens=4096) CSS = """ body { background: #000000 !important; } .gradio-container { background: #000000 !important; color: #e6edf3 !important; } .tabitem { background: #0a0a0a !important; border: 1px solid #1a1a1a !important; border-radius: 8px !important; } .tab-nav { background: #000000 !important; border-bottom: 2px solid #FF6B00 !important; } .tab-nav button { color: #888 !important; background: transparent !important; font-family: 'Roboto Mono', monospace !important; font-size: 0.85em !important; } .tab-nav button.selected { color: #FF6B00 !important; border-bottom: 2px solid #FF6B00 !important; font-weight: bold !important; } input, textarea, select { background: #111 !important; color: #00D4FF !important; border: 1px solid #333 !important; font-family: 'Roboto Mono', monospace !important; } button.primary { background: #FF6B00 !important; color: #000 !important; font-weight: 700 !important; font-family: 'Roboto Mono', monospace !important; border-radius: 4px !important; } button.secondary { background: #1a1a1a !important; color: #FF6B00 !important; border: 1px solid #FF6B00 !important; font-family: 'Roboto Mono', monospace !important; } .markdown-body { color: #e6edf3 !important; font-family: 'Roboto Mono', monospace !important; } .markdown-body h1 { color: #FF6B00 !important; border-bottom: 1px solid #333 !important; font-size: 1.3em !important; } .markdown-body h2 { color: #00D4FF !important; font-size: 1.1em !important; } .markdown-body h3 { color: #00C853 !important; font-size: 1em !important; } .markdown-body table { border-color: #333 !important; font-size: 0.85em !important; } .markdown-body code { background: #1a1a1a !important; color: #00D4FF !important; padding: 2px 6px !important; border-radius: 4px !important; } """ def build_app(): with gr.Blocks( title="AlphaForge V3.2 - Institutional Quant Platform", theme=gr.themes.Soft(primary_hue="orange", secondary_hue="cyan", neutral_hue="gray", font=[gr.themes.GoogleFont("Roboto Mono"), "monospace"]), css=CSS ) as app: gr.Markdown("""

ALPHAFORGE V3.2

Institutional Quant Trading Platform | K2 Think V2 Powered

Multi-Market: US | EU | UK | DE | JP | CN/HK | IN | Crypto | Forex | Commodities | Indices

""") with gr.Tabs(): with gr.TabItem("Technical Analysis"): with gr.Row(): with gr.Column(scale=1): ta_ticker = gr.Textbox(label="Ticker", value="AAPL") ta_market = gr.Dropdown(label="Market", choices=list(MARKETS.keys()), value="US Equities") ta_period = gr.Dropdown(label="Period", choices=["1mo","3mo","6mo","1y","2y","5y"], value="1y") ta_btn = gr.Button("Analyze") gr.Markdown("Examples: `AAPL` (US), `AIR.PA` (EU), `7203.T` (JP), `BTC-USD` (Crypto)") with gr.Column(scale=3): ta_out1 = gr.Plot() ta_out2 = gr.Plot() ta_out3 = gr.Plot() with gr.Row(): ta_out4 = gr.Plot() ta_out5 = gr.Plot() ta_out6 = gr.Plot() ta_md = gr.Markdown() ta_btn.click(fn=tech_analysis, inputs=[ta_ticker, ta_market, ta_period], outputs=[ta_out1, ta_out2, ta_out3, ta_out4, ta_out5, ta_out6, ta_md]) with gr.TabItem("AI Analysis (K2)"): with gr.Row(): with gr.Column(scale=1): ai_ticker = gr.Textbox(label="Ticker", value="AAPL") ai_market = gr.Dropdown(label="Market", choices=list(MARKETS.keys()), value="US Equities") ai_period = gr.Dropdown(label="Period", choices=["1mo","3mo","6mo","1y","2y"], value="1y") ai_btn = gr.Button("Generate AI Report") with gr.Column(scale=3): ai_out = gr.Textbox(label="K2 Think V2 Analysis", lines=30) ai_btn.click(fn=ai_analysis, inputs=[ai_ticker, ai_market, ai_period], outputs=ai_out) with gr.TabItem("Backtest"): with gr.Row(): with gr.Column(scale=1): bt_ticker = gr.Textbox(label="Ticker", value="AAPL") bt_strategy = gr.Dropdown(label="Strategy", choices=["Moving Average Crossover","RSI Strategy","MACD Momentum","Mean Reversion","Bollinger Squeeze"], value="Moving Average Crossover") bt_capital = gr.Number(label="Start Capital", value=100000) bt_risk = gr.Slider(label="Risk % per Trade", minimum=1, maximum=10, value=2, step=0.5) bt_period = gr.Dropdown(label="Period", choices=["1y","2y","5y"], value="2y") bt_btn = gr.Button("Run Backtest") with gr.Column(scale=3): bt_eq = gr.Plot() bt_dd = gr.Plot() with gr.Row(): bt_trades = gr.Dataframe() bt_md = gr.Markdown() bt_btn.click(fn=backtest, inputs=[bt_ticker, bt_strategy, bt_capital, bt_risk, bt_period], outputs=[bt_eq, bt_dd, bt_trades, bt_md, gr.Textbox(visible=False)]) with gr.TabItem("Portfolio Optimizer"): with gr.Row(): with gr.Column(scale=1): po_tickers = gr.Textbox(label="Tickers (comma-separated)", value="AAPL, MSFT, GOOGL, AMZN, NVDA") po_period = gr.Dropdown(label="Period", choices=["6mo","1y","2y"], value="1y") po_btn = gr.Button("Optimize Portfolio") with gr.Column(scale=3): po_frontier = gr.Plot() po_pie = gr.Plot() with gr.Row(): po_weights = gr.Dataframe() po_md = gr.Markdown() po_btn.click(fn=optimize_portfolio, inputs=[po_tickers, po_period], outputs=[po_frontier, po_pie, po_weights, po_md]) with gr.TabItem("Options Pricing"): with gr.Row(): with gr.Column(scale=1): op_ticker = gr.Textbox(label="Ticker", value="AAPL") op_type = gr.Dropdown(label="Option Type", choices=["call","put"], value="call") op_strike = gr.Slider(label="Strike % of Spot", minimum=50, maximum=150, value=100, step=1) op_days = gr.Slider(label="Days to Expiry", minimum=7, maximum=365, value=30, step=1) op_rfr = gr.Slider(label="Risk-Free Rate %", minimum=0, maximum=10, value=4.5, step=0.1) op_vol = gr.Slider(label="Vol Override % (0=auto)", minimum=0, maximum=100, value=0, step=1) op_btn = gr.Button("Price Option") with gr.Column(scale=3): op_greeks = gr.Plot() with gr.Row(): op_scenarios = gr.Dataframe() op_md = gr.Markdown() op_btn.click(fn=options_pricing, inputs=[op_ticker, op_strike, op_days, op_rfr, op_vol, op_type], outputs=[op_greeks, op_scenarios, op_md]) with gr.TabItem("Pairs Trading"): with gr.Row(): with gr.Column(scale=1): pt_a = gr.Textbox(label="Asset A", value="AAPL") pt_b = gr.Textbox(label="Asset B", value="MSFT") pt_period = gr.Dropdown(label="Period", choices=["6mo","1y","2y"], value="1y") pt_btn = gr.Button("Analyze Pair") with gr.Column(scale=3): pt_fig = gr.Plot() pt_scat = gr.Plot() pt_md = gr.Markdown() pt_btn.click(fn=pairs_trade, inputs=[pt_a, pt_b, pt_period], outputs=[pt_fig, pt_scat, pt_md]) with gr.TabItem("Crypto Arbitrage"): with gr.Row(): with gr.Column(scale=1): ca_coins = gr.Textbox(label="Coins (comma-separated)", value="BTC, ETH, SOL, XRP") ca_btn = gr.Button("Scan Arbitrage") with gr.Column(scale=3): ca_heatmap = gr.Plot() ca_md = gr.Markdown() ca_btn.click(fn=crypto_arbitrage, inputs=ca_coins, outputs=[ca_heatmap, ca_md]) with gr.TabItem("Risk Engine"): with gr.Row(): with gr.Column(scale=1): re_tickers = gr.Textbox(label="Tickers (comma-separated)", value="AAPL, MSFT, GOOGL, AMZN") re_stress = gr.Textbox(label="Stress Shocks JSON", value='{"AAPL":-10, "AMZN":5}', placeholder='{"AAPL":-10, "TSLA":15}') re_btn = gr.Button("Run Risk Analysis") with gr.Column(scale=3): re_corr = gr.Plot() re_dist = gr.Plot() re_md = gr.Markdown() re_btn.click(fn=risk_engine, inputs=[re_tickers, re_stress], outputs=[re_corr, re_dist, re_md]) with gr.TabItem("Sentiment"): with gr.Row(): with gr.Column(scale=1): se_ticker = gr.Textbox(label="Ticker", value="AAPL") se_btn = gr.Button("Analyze Sentiment") with gr.Column(scale=3): se_gauge = gr.Plot() se_md = gr.Markdown() se_btn.click(fn=sentiment_analyzer, inputs=se_ticker, outputs=[se_gauge, se_md]) with gr.TabItem("Macro"): with gr.Row(): with gr.Column(scale=1): ma_btn = gr.Button("Refresh Macro Dashboard") with gr.Column(scale=3): ma_fig = gr.Plot() ma_md = gr.Markdown() ma_btn.click(fn=macro_analysis, inputs=[], outputs=[ma_fig, ma_md]) with gr.TabItem("Chat (K2)"): with gr.Row(): with gr.Column(scale=1): chat_input = gr.Textbox(label="Ask K2 Think V2", value="Explain gamma scalping", lines=3) chat_btn = gr.Button("Send") with gr.Column(scale=3): chat_output = gr.Textbox(label="Response", lines=30) chat_btn.click( fn=lambda q: K2ThinkClient().chat([{"role":"user","content":q}], temperature=0.4, max_tokens=4096), inputs=chat_input, outputs=chat_output ) gr.Markdown("""

AlphaForge V3.2 | For demonstration purposes only. Not financial advice.
Configure K2_API_KEY in Space Settings > Repository Secrets for AI features.

""") return app if __name__ == "__main__": app = build_app() app.launch()