diff --git a/package.json b/package.json
index 0274dc1..fc3a3f9 100644
--- a/package.json
+++ b/package.json
@@ -1,6 +1,6 @@
 {
   "name": "9router-app",
-  "version": "0.3.73",
+  "version": "0.3.74",
   "description": "9Router web dashboard",
   "private": true,
   "scripts": {
diff --git a/src/lib/localDb.js b/src/lib/localDb.js
index 00983b6..aa81fb5 100644
--- a/src/lib/localDb.js
+++ b/src/lib/localDb.js
@@ -1080,93 +1080,65 @@ export async function getCloudUrl() {
 
 /**
  * Get pricing configuration
- * Returns merged user pricing with defaults
+ * Returns merged: PROVIDER_PRICING defaults + user overrides
  */
 export async function getPricing() {
   const db = await getDb();
   const userPricing = db.data.pricing || {};
 
-  // Import default pricing
-  const { getDefaultPricing } = await import("@/shared/constants/pricing.js");
-  const defaultPricing = getDefaultPricing();
+  const { PROVIDER_PRICING } = await import("@/shared/constants/pricing.js");
 
-  // Merge user pricing with defaults
-  // User pricing overrides defaults for specific provider/model combinations
-  const mergedPricing = {};
+  // Deep merge PROVIDER_PRICING + user overrides
+  const merged = {};
 
-  for (const [provider, models] of Object.entries(defaultPricing)) {
-    mergedPricing[provider] = { ...models };
-
-    // Apply user overrides if they exist
+  for (const [provider, models] of Object.entries(PROVIDER_PRICING)) {
+    merged[provider] = { ...models };
     if (userPricing[provider]) {
       for (const [model, pricing] of Object.entries(userPricing[provider])) {
-        if (mergedPricing[provider][model]) {
-          mergedPricing[provider][model] = { ...mergedPricing[provider][model], ...pricing };
-        } else {
-          mergedPricing[provider][model] = pricing;
-        }
+        merged[provider][model] = merged[provider][model]
+          ? { ...merged[provider][model], ...pricing }
+          : pricing;
       }
     }
   }
 
-  // Add any user-only pricing entries
+  // User-only providers not in PROVIDER_PRICING
   for (const [provider, models] of Object.entries(userPricing)) {
-    if (!mergedPricing[provider]) {
-      mergedPricing[provider] = { ...models };
+    if (!merged[provider]) {
+      merged[provider] = { ...models };
     } else {
       for (const [model, pricing] of Object.entries(models)) {
-        if (!mergedPricing[provider][model]) {
-          mergedPricing[provider][model] = pricing;
-        }
+        if (!merged[provider][model]) merged[provider][model] = pricing;
       }
     }
   }
 
-  return mergedPricing;
+  return merged;
 }
 
 /**
- * Get pricing for a specific provider and model
+ * Get pricing for a specific provider and model.
+ * Delegates to getPricingForModel in pricing.js which handles the full fallback chain:
+ *   1. PROVIDER_PRICING[provider][model]  — provider-specific override
+ *   2. MODEL_PRICING[model]               — canonical model price
+ *   3. PATTERN_PRICING                    — glob pattern match
+ *
+ * Also checks user DB overrides first.
  */
 export async function getPricingForModel(provider, model) {
-  const pricing = await getPricing();
+  if (!model) return null;
 
-  // Try direct lookup
-  if (pricing[provider]?.[model]) {
-    return pricing[provider][model];
+  const db = await getDb();
+  const userPricing = db.data.pricing || {};
+
+  // User override takes top priority
+  if (provider && userPricing[provider]?.[model]) {
+    return userPricing[provider][model];
   }
 
-  // Try mapping provider ID to alias
-  // We need to duplicate the mapping here or import it
-  // Since we can't easily import from open-sse, we'll implement the mapping locally
-  const PROVIDER_ID_TO_ALIAS = {
-    claude: "cc",
-    codex: "cx",
-    "gemini-cli": "gc",
-    qwen: "qw",
-    iflow: "if",
-    antigravity: "ag",
-    github: "gh",
-    kiro: "kr",
-    openai: "openai",
-    anthropic: "anthropic",
-    gemini: "gemini",
-    openrouter: "openrouter",
-    glm: "glm",
-    kimi: "kimi",
-    minimax: "minimax",
-  };
-
-  const alias = PROVIDER_ID_TO_ALIAS[provider];
-  if (alias && pricing[alias]) {
-    return pricing[alias][model] || null;
-  }
-
-  // Fallback: strip vendor prefix (e.g. "deepseek/deepseek-chat" → "deepseek-chat")
-  // then lookup in MODEL_PRICING (provider-agnostic explicit map)
-  const { MODEL_PRICING } = await import("@/shared/constants/pricing.js");
-  const baseModel = model.includes("/") ? model.split("/").pop() : model;
-  return MODEL_PRICING[baseModel] || MODEL_PRICING[model] || null;
+  // Delegate to constants fallback chain
+  const { getPricingForModel: resolve } = await import("@/shared/constants/pricing.js");
+  return resolve(provider, model);
 }
 
 /**
diff --git a/src/lib/usageDb.js b/src/lib/usageDb.js
index c3a6a3c..f6d24bc 100644
--- a/src/lib/usageDb.js
+++ b/src/lib/usageDb.js
@@ -92,6 +92,12 @@ if (!global._statsEmitter) {
 }
 export const statsEmitter = global._statsEmitter;
 
+// Safety timers — force-clear pending counts after 1 min if END was never called
+if (!global._pendingTimers) global._pendingTimers = {};
+const pendingTimers = global._pendingTimers;
+
+const PENDING_TIMEOUT_MS = 60 * 1000; // 1 minute
+
 /**
  * Track a pending request
  * @param {string} model
@@ -102,6 +108,7 @@ export const statsEmitter = global._statsEmitter;
  */
 export function trackPendingRequest(model, provider, connectionId, started, error = false) {
   const modelKey = provider ? `${model} (${provider})` : model;
+  const timerKey = `${connectionId}|${modelKey}`;
 
   // Track by model
   if (!pendingRequests.byModel[modelKey]) pendingRequests.byModel[modelKey] = 0;
@@ -109,10 +116,28 @@ export function trackPendingRequest(model, provider, connectionId, started, erro
 
   // Track by account
   if (connectionId) {
-    const accountKey = connectionId;
-    if (!pendingRequests.byAccount[accountKey]) pendingRequests.byAccount[accountKey] = {};
-    if (!pendingRequests.byAccount[accountKey][modelKey]) pendingRequests.byAccount[accountKey][modelKey] = 0;
-    pendingRequests.byAccount[accountKey][modelKey] = Math.max(0, pendingRequests.byAccount[accountKey][modelKey] + (started ? 1 : -1));
+    if (!pendingRequests.byAccount[connectionId]) pendingRequests.byAccount[connectionId] = {};
+    if (!pendingRequests.byAccount[connectionId][modelKey]) pendingRequests.byAccount[connectionId][modelKey] = 0;
+    pendingRequests.byAccount[connectionId][modelKey] = Math.max(0, pendingRequests.byAccount[connectionId][modelKey] + (started ? 1 : -1));
+  }
+
+  if (started) {
+    // Safety timeout: force-clear if END is never called (client disconnect, crash, etc.)
+    clearTimeout(pendingTimers[timerKey]);
+    pendingTimers[timerKey] = setTimeout(() => {
+      delete pendingTimers[timerKey];
+      if (pendingRequests.byModel[modelKey] > 0) {
+        pendingRequests.byModel[modelKey] = 0;
+      }
+      if (connectionId && pendingRequests.byAccount[connectionId]?.[modelKey] > 0) {
+        pendingRequests.byAccount[connectionId][modelKey] = 0;
+      }
+      statsEmitter.emit("pending");
+    }, PENDING_TIMEOUT_MS);
+  } else {
+    // END called normally — cancel the safety timer
+    clearTimeout(pendingTimers[timerKey]);
+    delete pendingTimers[timerKey];
   }
 
   // Track error provider (auto-clears after 10s)
diff --git a/src/shared/components/NineRemoteButton.js b/src/shared/components/NineRemoteButton.js
index 8c2d1af..c71ed8e 100644
--- a/src/shared/components/NineRemoteButton.js
+++ b/src/shared/components/NineRemoteButton.js
@@ -1,104 +1,23 @@
 "use client";
 
-import { useState, useEffect, useRef } from "react";
-import NineRemoteModal from "./NineRemoteModal";
-
-// step 0-4 từ 9remote SSE: 0=Stopped,1=Preparing,2=Connecting,3=Tunneling,4=Ready
-const STEP = { STOPPED: 0, PREPARING: 1, CONNECTING: 2, TUNNELING: 3, READY: 4 };
-
-// Retry interval khi 9remote chưa chạy (30s để không spam)
-const RETRY_MS = 30000;
-
-const stepStyle = (step, installed) => {
-  if (!installed) return { color: "text-text-muted hover:text-text-main hover:bg-black/5 dark:hover:bg-white/5", glow: null };
-  switch (step) {
-    case STEP.READY:   return { color: "text-emerald-500 hover:bg-emerald-500/10", glow: "drop-shadow(0 0 6px rgb(16 185 129 / 0.7))" };
-    case STEP.STOPPED: return { color: "text-text-muted hover:text-text-main hover:bg-black/5 dark:hover:bg-white/5", glow: null };
-    default:           return { color: "text-amber-400 hover:bg-amber-400/10", glow: null };
-  }
-};
+import { useState } from "react";
+import NineRemotePromoModal from "./NineRemotePromoModal";
 
 export default function NineRemoteButton() {
   const [isOpen, setIsOpen] = useState(false);
-  const [installed, setInstalled] = useState(false);
-  const [step, setStep] = useState(STEP.STOPPED);
-  const esRef = useRef(null);
-  const retryRef = useRef(null);
-
-  const scheduleRetry = () => {
-    clearTimeout(retryRef.current);
-    retryRef.current = setTimeout(connect, RETRY_MS);
-  };
-
-  const connect = () => {
-    esRef.current?.close();
-    clearTimeout(retryRef.current);
-
-    const es = new EventSource("http://localhost:2208/api/ui/events");
-
-    es.onmessage = (e) => {
-      try {
-        const data = JSON.parse(e.data);
-        if (data.type === "state") {
-          setInstalled(true);
-          setStep(data.step ?? STEP.STOPPED);
-        }
-      } catch {}
-    };
-
-    es.onerror = () => {
-      es.close();
-      setStep(STEP.STOPPED);
-      // 9remote not running — retry after delay
-      scheduleRetry();
-    };
-
-    esRef.current = es;
-  };
-
-  useEffect(() => {
-    // Check installed once on mount (no polling, just 1 call)
-    fetch("/api/9remote/status")
-      .then((r) => r.json())
-      .then((d) => setInstalled(d.installed))
-      .catch(() => {});
-
-    connect();
-
-    return () => {
-      esRef.current?.close();
-      clearTimeout(retryRef.current);
-    };
-  }, []);
-
-  // When modal closes, reconnect SSE immediately (user may have just started 9remote)
-  const handleClose = () => {
-    setIsOpen(false);
-    setTimeout(connect, 800);
-  };
-
-  const { color, glow } = stepStyle(step, installed);
-  const isPulsing = installed && step >= STEP.PREPARING && step <= STEP.TUNNELING;
 
   return (
     <>
       <button
         onClick={() => setIsOpen(true)}
-        className={`relative flex items-center gap-1.5 px-2.5 py-1.5 rounded-lg transition-all ${color}`}
-        style={glow ? { filter: glow } : {}}
+        className="relative flex items-center gap-1.5 px-2.5 py-1.5 rounded-lg transition-all text-text-muted hover:text-text-main hover:bg-black/5 dark:hover:bg-white/5"
         title="9Remote"
       >
-        <span className={`material-symbols-outlined text-[18px]${isPulsing ? " animate-pulse" : ""}`}>
-          computer
-        </span>
+        <span className="material-symbols-outlined text-[18px]">computer</span>
         <span className="text-xs font-medium">Remote</span>
       </button>
 
-      <NineRemoteModal
-        isOpen={isOpen}
-        onClose={handleClose}
-        onInstalled={() => setInstalled(true)}
-      />
+      <NineRemotePromoModal isOpen={isOpen} onClose={() => setIsOpen(false)} />
     </>
   );
 }
diff --git a/src/shared/components/NineRemotePromoModal.js b/src/shared/components/NineRemotePromoModal.js
new file mode 100644
index 0000000..7fc4cb8
--- /dev/null
+++ b/src/shared/components/NineRemotePromoModal.js
@@ -0,0 +1,103 @@
+"use client";
+
+import { useEffect } from "react";
+import { createPortal } from "react-dom";
+
+const FEATURES = [
+  { icon: "terminal", label: "Terminal", desc: "Full shell access" },
+  { icon: "cast", label: "Desktop", desc: "Screen sharing" },
+  { icon: "folder_open", label: "Files", desc: "Browse & edit files" },
+];
+
+const BULLETS = [
+  { icon: "qr_code_scanner", text: "Scan QR to connect instantly" },
+  { icon: "wifi_off", text: "No port forwarding needed" },
+  { icon: "devices", text: "Works on any device" },
+];
+
+const NINE_REMOTE_URL = "https://9remote.cc";
+
+export default function NineRemotePromoModal({ isOpen, onClose }) {
+  useEffect(() => {
+    if (!isOpen) return;
+    document.body.style.overflow = "hidden";
+    const onEsc = (e) => { if (e.key === "Escape") onClose(); };
+    document.addEventListener("keydown", onEsc);
+    return () => { document.body.style.overflow = ""; document.removeEventListener("keydown", onEsc); };
+  }, [isOpen, onClose]);
+
+  if (!isOpen) return null;
+
+  return createPortal(
+    <div className="fixed inset-0 z-[9999] flex items-center justify-center p-4">
+      <div className="absolute inset-0 bg-black/60 backdrop-blur-sm" onClick={onClose} />
+
+      <div className="relative w-full max-w-sm rounded-xl overflow-hidden shadow-2xl animate-in fade-in zoom-in-95 duration-200 flex flex-col bg-surface border border-black/10 dark:border-white/10">
+        {/* Header */}
+        <div className="flex items-center justify-between px-5 py-4 border-b border-black/5 dark:border-white/5">
+          <div className="flex items-center gap-3">
+            <div className="w-7 h-7 rounded-lg flex items-center justify-center" style={{ background: "#FF570A" }}>
+              <span className="material-symbols-outlined text-white text-base">terminal</span>
+            </div>
+            <span className="text-xs font-bold uppercase tracking-wider" style={{ fontFamily: "monospace", color: "#FF570A" }}>9Remote</span>
+          </div>
+          <button
+            onClick={onClose}
+            className="w-7 h-7 flex items-center justify-center rounded-lg bg-black/5 dark:bg-white/5 border border-black/5 dark:border-white/10 text-text-muted hover:text-text-main transition-colors"
+          >
+            <span className="material-symbols-outlined text-base">close</span>
+          </button>
+        </div>
+
+        {/* Body */}
+        <div className="px-7 py-7 pb-9 flex flex-col gap-6">
+          {/* Hero */}
+          <div className="flex flex-col items-center gap-2 text-center mt-2">
+            <div
+              className="w-14 h-14 rounded-2xl flex items-center justify-center mb-1"
+              style={{ background: "#FF570A", boxShadow: "rgba(255,87,10,0.35) 0px 8px 32px" }}
+            >
+              <span className="material-symbols-outlined text-white" style={{ fontSize: 30 }}>terminal</span>
+            </div>
+            <h1 className="text-lg font-bold text-text-main tracking-tight">9Remote</h1>
+            <p className="text-xs text-text-muted leading-5 max-w-[220px]">
+              Access your terminal, desktop &amp; files from anywhere
+            </p>
+          </div>
+
+          {/* Feature cards */}
+          <div className="flex gap-2 w-full">
+            {FEATURES.map(({ icon, label, desc }) => (
+              <div key={label} className="flex-1 flex flex-col items-center gap-1.5 py-4 px-1 rounded-xl border border-black/10 dark:border-white/10 bg-bg-alt">
+                <span className="material-symbols-outlined" style={{ fontSize: 22, color: "#ff6e33" }}>{icon}</span>
+                <p className="text-xs font-semibold text-text-main">{label}</p>
+                <p className="text-[10px] text-text-muted text-center leading-4">{desc}</p>
+              </div>
+            ))}
+          </div>
+
+          {/* Bullets */}
+          <div className="flex flex-col gap-3 w-full">
+            {BULLETS.map(({ icon, text }) => (
+              <div key={icon} className="flex items-center gap-2.5">
+                <span className="material-symbols-outlined flex-shrink-0" style={{ fontSize: 16, color: "#ff6e33" }}>{icon}</span>
+                <span className="text-xs text-text-muted">{text}</span>
+              </div>
+            ))}
+          </div>
+
+          {/* CTA */}
+          <button
+            onClick={() => window.open(NINE_REMOTE_URL, "_blank")}
+            className="w-full py-3.5 flex items-center justify-center gap-2 text-sm font-semibold text-white rounded-xl hover:opacity-90 active:scale-[0.98] transition-all"
+            style={{ background: "#FF570A", boxShadow: "0 4px 16px rgba(255,87,10,0.35)" }}
+          >
+            <span className="material-symbols-outlined text-base">open_in_new</span>
+            Get 9Remote
+          </button>
+        </div>
+      </div>
+    </div>,
+    document.body
+  );
+}
diff --git a/src/shared/constants/pricing.js b/src/shared/constants/pricing.js
index f96038e..8df1bf6 100644
--- a/src/shared/constants/pricing.js
+++ b/src/shared/constants/pricing.js
@@ -1,829 +1,257 @@
-// Default pricing rates for AI models
-// All rates are in dollars per million tokens ($/1M tokens)
-// Based on user-provided pricing for Antigravity models and industry standards for others
-
-export const DEFAULT_PRICING = {
-  // OAuth Providers (using aliases)
-
-  // Claude Code (cc)
-  cc: {
-    "claude-opus-4-6": {
-      input: 5.00,
-      output: 25.00,
-      cached: 0.50,
-      reasoning: 25.00,
-      cache_creation: 6.25
-    },
-    "claude-opus-4-5-20251101": {
-      input: 5.00,
-      output: 25.00,
-      cached: 0.50,
-      reasoning: 25.00,
-      cache_creation: 6.25
-    },
-    "claude-sonnet-4-6": {
-      input: 3.00,
-      output: 15.00,
-      cached: 0.30,
-      reasoning: 15.00,
-      cache_creation: 3.75
-    },
-    "claude-sonnet-4-5-20250929": {
-      input: 3.00,
-      output: 15.00,
-      cached: 0.30,
-      reasoning: 15.00,
-      cache_creation: 3.75
-    },
-    "claude-haiku-4-5-20251001": {
-      input: 1.00,
-      output: 5.00,
-      cached: 0.10,
-      reasoning: 5.00,
-      cache_creation: 1.25
-    }
-  },
-
-  // OpenAI Codex (cx)
-  cx: {
-    "gpt-5.3-codex": {
-      input: 6.00,
-      output: 24.00,
-      cached: 3.00,
-      reasoning: 36.00,
-      cache_creation: 6.00
-    },
-    "gpt-5.3-codex-xhigh": {
-      input: 10.00,
-      output: 40.00,
-      cached: 5.00,
-      reasoning: 60.00,
-      cache_creation: 10.00
-    },
-    "gpt-5.3-codex-high": {
-      input: 8.00,
-      output: 32.00,
-      cached: 4.00,
-      reasoning: 48.00,
-      cache_creation: 8.00
-    },
-    "gpt-5.3-codex-low": {
-      input: 4.00,
-      output: 16.00,
-      cached: 2.00,
-      reasoning: 24.00,
-      cache_creation: 4.00
-    },
-    "gpt-5.3-codex-none": {
-      input: 3.00,
-      output: 12.00,
-      cached: 1.50,
-      reasoning: 18.00,
-      cache_creation: 3.00
-    },
-    "gpt-5.3-codex-spark": {
-      input: 3.00,
-      output: 12.00,
-      cached: 0.30,
-      reasoning: 12.00,
-      cache_creation: 3.00
-    },
-    "gpt-5.2-codex": {
-      input: 5.00,
-      output: 20.00,
-      cached: 2.50,
-      reasoning: 30.00,
-      cache_creation: 5.00
-    },
-    "gpt-5.2": {
-      input: 5.00,
-      output: 20.00,
-      cached: 2.50,
-      reasoning: 30.00,
-      cache_creation: 5.00
-    },
-    "gpt-5.1-codex-max": {
-      input: 8.00,
-      output: 32.00,
-      cached: 4.00,
-      reasoning: 48.00,
-      cache_creation: 8.00
-    },
-    "gpt-5.1-codex": {
-      input: 4.00,
-      output: 16.00,
-      cached: 2.00,
-      reasoning: 24.00,
-      cache_creation: 4.00
-    },
-    "gpt-5.1-codex-mini": {
-      input: 1.50,
-      output: 6.00,
-      cached: 0.75,
-      reasoning: 9.00,
-      cache_creation: 1.50
-    },
-    "gpt-5.1-codex-mini-high": {
-      input: 2.00,
-      output: 8.00,
-      cached: 1.00,
-      reasoning: 12.00,
-      cache_creation: 2.00
-    },
-    "gpt-5.1": {
-      input: 4.00,
-      output: 16.00,
-      cached: 2.00,
-      reasoning: 24.00,
-      cache_creation: 4.00
-    },
-    "gpt-5-codex": {
-      input: 3.00,
-      output: 12.00,
-      cached: 1.50,
-      reasoning: 18.00,
-      cache_creation: 3.00
-    },
-    "gpt-5-codex-mini": {
-      input: 1.00,
-      output: 4.00,
-      cached: 0.50,
-      reasoning: 6.00,
-      cache_creation: 1.00
-    }
-  },
-
-  // Gemini CLI (gc)
-  gc: {
-    "gemini-3-flash-preview": {
-      input: 0.50,
-      output: 3.00,
-      cached: 0.03,
-      reasoning: 4.50,
-      cache_creation: 0.50
-    },
-    "gemini-3-pro-preview": {
-      input: 2.00,
-      output: 12.00,
-      cached: 0.25,
-      reasoning: 18.00,
-      cache_creation: 2.00
-    },
-    "gemini-2.5-pro": {
-      input: 2.00,
-      output: 12.00,
-      cached: 0.25,
-      reasoning: 18.00,
-      cache_creation: 2.00
-    },
-    "gemini-2.5-flash": {
-      input: 0.30,
-      output: 2.50,
-      cached: 0.03,
-      reasoning: 3.75,
-      cache_creation: 0.30
-    },
-    "gemini-2.5-flash-lite": {
-      input: 0.15,
-      output: 1.25,
-      cached: 0.015,
-      reasoning: 1.875,
-      cache_creation: 0.15
-    }
-  },
-
-  // Qwen Code (qw)
-  qw: {
-    "qwen3-coder-plus": {
-      input: 1.00,
-      output: 4.00,
-      cached: 0.50,
-      reasoning: 6.00,
-      cache_creation: 1.00
-    },
-    "qwen3-coder-flash": {
-      input: 0.50,
-      output: 2.00,
-      cached: 0.25,
-      reasoning: 3.00,
-      cache_creation: 0.50
-    },
-    "vision-model": {
-      input: 1.50,
-      output: 6.00,
-      cached: 0.75,
-      reasoning: 9.00,
-      cache_creation: 1.50
-    },
-    "coder-model": {
-      input: 1.50,
-      output: 6.00,
-      cached: 0.75,
-      reasoning: 9.00,
-      cache_creation: 1.50
-    }
-  },
-
-  // iFlow AI (if)
-  if: {
-    "qwen3-coder-plus": {
-      input: 1.00,
-      output: 4.00,
-      cached: 0.50,
-      reasoning: 6.00,
-      cache_creation: 1.00
-    },
-    "kimi-k2": {
-      input: 1.00,
-      output: 4.00,
-      cached: 0.50,
-      reasoning: 6.00,
-      cache_creation: 1.00
-    },
-    "kimi-k2-thinking": {
-      input: 1.50,
-      output: 6.00,
-      cached: 0.75,
-      reasoning: 9.00,
-      cache_creation: 1.50
-    },
-    "kimi-k2.5": {
-      input: 1.20,
-      output: 4.80,
-      cached: 0.60,
-      reasoning: 7.20,
-      cache_creation: 1.20
-    },
-    "deepseek-r1": {
-      input: 0.75,
-      output: 3.00,
-      cached: 0.375,
-      reasoning: 4.50,
-      cache_creation: 0.75
-    },
-    "deepseek-v3.2-chat": {
-      input: 0.50,
-      output: 2.00,
-      cached: 0.25,
-      reasoning: 3.00,
-      cache_creation: 0.50
-    },
-    "deepseek-v3.2-reasoner": {
-      input: 0.75,
-      output: 3.00,
-      cached: 0.375,
-      reasoning: 4.50,
-      cache_creation: 0.75
-    },
-    "minimax-m2.1": {
-      input: 0.50,
-      output: 2.00,
-      cached: 0.25,
-      reasoning: 3.00,
-      cache_creation: 0.50
-    },
-    "minimax-m2.5": {
-      input: 0.60,
-      output: 2.40,
-      cached: 0.30,
-      reasoning: 3.60,
-      cache_creation: 0.60
-    },
-    "glm-4.6": {
-      input: 0.50,
-      output: 2.00,
-      cached: 0.25,
-      reasoning: 3.00,
-      cache_creation: 0.50
-    },
-    "glm-4.7": {
-      input: 0.75,
-      output: 3.00,
-      cached: 0.375,
-      reasoning: 4.50,
-      cache_creation: 0.75
-    },
-    "glm-5": {
-      input: 1.00,
-      output: 4.00,
-      cached: 0.50,
-      reasoning: 6.00,
-      cache_creation: 1.00
-    }
-  },
-
-  // Antigravity (ag) - User-provided pricing
-  ag: {
-    "gemini-3.1-pro-low": {
-      input: 2.00,
-      output: 12.00,
-      cached: 0.25,
-      reasoning: 18.00,
-      cache_creation: 2.00
-    },
-    "gemini-3.1-pro-high": {
-      input: 4.00,
-      output: 18.00,
-      cached: 0.50,
-      reasoning: 27.00,
-      cache_creation: 4.00
-    },
-    "gemini-3-flash": {
-      input: 0.50,
-      output: 3.00,
-      cached: 0.03,
-      reasoning: 4.50,
-      cache_creation: 0.50
-    },
-    "gemini-2.5-flash": {
-      input: 0.30,
-      output: 2.50,
-      cached: 0.03,
-      reasoning: 3.75,
-      cache_creation: 0.30
-    },
-    "claude-sonnet-4-6": {
-      input: 3.00,
-      output: 15.00,
-      cached: 0.30,
-      reasoning: 22.50,
-      cache_creation: 3.00
-    },
-    "claude-opus-4-5-thinking": {
-      input: 5.00,
-      output: 25.00,
-      cached: 0.50,
-      reasoning: 37.50,
-      cache_creation: 5.00
-    },
-    "claude-opus-4-6-thinking": {
-      input: 5.00,
-      output: 25.00,
-      cached: 0.50,
-      reasoning: 37.50,
-      cache_creation: 5.00
-    },
-    "gpt-oss-120b-medium": {
-      input: 0.50,
-      output: 2.00,
-      cached: 0.25,
-      reasoning: 3.00,
-      cache_creation: 0.50
-    }
-  },
-
-  // GitHub Copilot (gh)
-  gh: {
-    "gpt-3.5-turbo": {
-      input: 0.50,
-      output: 1.50,
-      cached: 0.25,
-      reasoning: 2.25,
-      cache_creation: 0.50
-    },
-    "gpt-4": {
-      input: 2.50,
-      output: 10.00,
-      cached: 1.25,
-      reasoning: 15.00,
-      cache_creation: 2.50
-    },
-    "gpt-4o": {
-      input: 2.50,
-      output: 10.00,
-      cached: 1.25,
-      reasoning: 15.00,
-      cache_creation: 2.50
-    },
-    "gpt-4o-mini": {
-      input: 0.15,
-      output: 0.60,
-      cached: 0.075,
-      reasoning: 0.90,
-      cache_creation: 0.15
-    },
-    "gpt-4.1": {
-      input: 2.50,
-      output: 10.00,
-      cached: 1.25,
-      reasoning: 15.00,
-      cache_creation: 2.50
-    },
-    "gpt-5": {
-      input: 3.00,
-      output: 12.00,
-      cached: 1.50,
-      reasoning: 18.00,
-      cache_creation: 3.00
-    },
-    "gpt-5-mini": {
-      input: 0.75,
-      output: 3.00,
-      cached: 0.375,
-      reasoning: 4.50,
-      cache_creation: 0.75
-    },
-    "gpt-5-codex": {
-      input: 3.00,
-      output: 12.00,
-      cached: 1.50,
-      reasoning: 18.00,
-      cache_creation: 3.00
-    },
-    "gpt-5.1": {
-      input: 4.00,
-      output: 16.00,
-      cached: 2.00,
-      reasoning: 24.00,
-      cache_creation: 4.00
-    },
-    "gpt-5.1-codex": {
-      input: 4.00,
-      output: 16.00,
-      cached: 2.00,
-      reasoning: 24.00,
-      cache_creation: 4.00
-    },
-    "gpt-5.1-codex-mini": {
-      input: 1.50,
-      output: 6.00,
-      cached: 0.75,
-      reasoning: 9.00,
-      cache_creation: 1.50
-    },
-    "gpt-5.1-codex-max": {
-      input: 8.00,
-      output: 32.00,
-      cached: 4.00,
-      reasoning: 48.00,
-      cache_creation: 8.00
-    },
-    "gpt-5.2": {
-      input: 5.00,
-      output: 20.00,
-      cached: 2.50,
-      reasoning: 30.00,
-      cache_creation: 5.00
-    },
-    "gpt-5.2-codex": {
-      input: 5.00,
-      output: 20.00,
-      cached: 2.50,
-      reasoning: 30.00,
-      cache_creation: 5.00
-    },
-    "gpt-5.3-codex": {
-      input: 1.75,
-      output: 14.00,
-      cached: 0.175,
-      reasoning: 14.00,
-      cache_creation: 1.75
-    },
-    "claude-haiku-4.5": {
-      input: 0.50,
-      output: 2.50,
-      cached: 0.05,
-      reasoning: 3.75,
-      cache_creation: 0.50
-    },
-    "claude-opus-4.1": {
-      input: 5.00,
-      output: 25.00,
-      cached: 0.50,
-      reasoning: 37.50,
-      cache_creation: 5.00
-    },
-    "claude-opus-4.5": {
-      input: 5.00,
-      output: 25.00,
-      cached: 0.50,
-      reasoning: 37.50,
-      cache_creation: 5.00
-    },
-    "claude-sonnet-4": {
-      input: 3.00,
-      output: 15.00,
-      cached: 0.30,
-      reasoning: 22.50,
-      cache_creation: 3.00
-    },
-    "claude-sonnet-4.5": {
-      input: 3.00,
-      output: 15.00,
-      cached: 0.30,
-      reasoning: 22.50,
-      cache_creation: 3.00
-    },
-    "claude-sonnet-4.6": {
-      input: 3.00,
-      output: 15.00,
-      cached: 0.30,
-      reasoning: 22.50,
-      cache_creation: 3.00
-    },
-    "claude-opus-4.6": {
-      input: 5.00,
-      output: 25.00,
-      cached: 0.50,
-      reasoning: 37.50,
-      cache_creation: 5.00
-    },
-    "gemini-2.5-pro": {
-      input: 2.00,
-      output: 12.00,
-      cached: 0.25,
-      reasoning: 18.00,
-      cache_creation: 2.00
-    },
-    "gemini-3-flash-preview": {
-      input: 0.50,
-      output: 3.00,
-      cached: 0.03,
-      reasoning: 4.50,
-      cache_creation: 0.50
-    },
-    "gemini-3-pro-preview": {
-      input: 2.00,
-      output: 12.00,
-      cached: 0.25,
-      reasoning: 18.00,
-      cache_creation: 2.00
-    },
-    "grok-code-fast-1": {
-      input: 0.50,
-      output: 2.00,
-      cached: 0.25,
-      reasoning: 3.00,
-      cache_creation: 0.50
-    },
-    "oswe-vscode-prime": {
-      input: 1.00,
-      output: 4.00,
-      cached: 0.50,
-      reasoning: 6.00,
-      cache_creation: 1.00
-    }
-  },
-
-  // Kiro AI (kr) - AWS CodeWhisperer
-  kr: {
-    "claude-sonnet-4.5": {
-      input: 3.00,
-      output: 15.00,
-      cached: 0.30,
-      reasoning: 22.50,
-      cache_creation: 3.00
-    },
-    "claude-haiku-4.5": {
-      input: 0.50,
-      output: 2.50,
-      cached: 0.05,
-      reasoning: 3.75,
-      cache_creation: 0.50
-    }
-  },
-
-  // API Key Providers (alias = id)
-
-  // OpenAI
-  openai: {
-    "gpt-4o": {
-      input: 2.50,
-      output: 10.00,
-      cached: 1.25,
-      reasoning: 15.00,
-      cache_creation: 2.50
-    },
-    "gpt-4o-mini": {
-      input: 0.15,
-      output: 0.60,
-      cached: 0.075,
-      reasoning: 0.90,
-      cache_creation: 0.15
-    },
-    "gpt-4-turbo": {
-      input: 10.00,
-      output: 30.00,
-      cached: 5.00,
-      reasoning: 45.00,
-      cache_creation: 10.00
-    },
-    "o1": {
-      input: 15.00,
-      output: 60.00,
-      cached: 7.50,
-      reasoning: 90.00,
-      cache_creation: 15.00
-    },
-    "o1-mini": {
-      input: 3.00,
-      output: 12.00,
-      cached: 1.50,
-      reasoning: 18.00,
-      cache_creation: 3.00
-    }
-  },
-
-  // Anthropic
-  anthropic: {
-    "claude-sonnet-4-20250514": {
-      input: 3.00,
-      output: 15.00,
-      cached: 1.50,
-      reasoning: 15.00,
-      cache_creation: 3.00
-    },
-    "claude-opus-4-20250514": {
-      input: 15.00,
-      output: 25.00,
-      cached: 7.50,
-      reasoning: 112.50,
-      cache_creation: 15.00
-    },
-    "claude-3-5-sonnet-20241022": {
-      input: 3.00,
-      output: 15.00,
-      cached: 1.50,
-      reasoning: 15.00,
-      cache_creation: 3.00
-    }
-  },
-
-  // Gemini
-  gemini: {
-    "gemini-3-pro-preview": {
-      input: 2.00,
-      output: 12.00,
-      cached: 0.25,
-      reasoning: 18.00,
-      cache_creation: 2.00
-    },
-    "gemini-2.5-pro": {
-      input: 2.00,
-      output: 12.00,
-      cached: 0.25,
-      reasoning: 18.00,
-      cache_creation: 2.00
-    },
-    "gemini-2.5-flash": {
-      input: 0.30,
-      output: 2.50,
-      cached: 0.03,
-      reasoning: 3.75,
-      cache_creation: 0.30
-    },
-    "gemini-2.5-flash-lite": {
-      input: 0.15,
-      output: 1.25,
-      cached: 0.015,
-      reasoning: 1.875,
-      cache_creation: 0.15
-    }
-  },
-
-  // OpenRouter
-  openrouter: {
-    "auto": {
-      input: 2.00,
-      output: 8.00,
-      cached: 1.00,
-      reasoning: 12.00,
-      cache_creation: 2.00
-    }
-  },
-
-  // GLM
-  glm: {
-    "glm-4.7": {
-      input: 0.75,
-      output: 3.00,
-      cached: 0.375,
-      reasoning: 4.50,
-      cache_creation: 0.75
-    },
-    "glm-4.6": {
-      input: 0.50,
-      output: 2.00,
-      cached: 0.25,
-      reasoning: 3.00,
-      cache_creation: 0.50
-    },
-    "glm-4.6v": {
-      input: 0.75,
-      output: 3.00,
-      cached: 0.375,
-      reasoning: 4.50,
-      cache_creation: 0.75
-    }
-  },
-
-  // Kimi
-  kimi: {
-    "kimi-k2.5": {
-      input: 1.20,
-      output: 4.80,
-      cached: 0.60,
-      reasoning: 7.20,
-      cache_creation: 1.20
-    },
-    "kimi-k2.5-thinking": {
-      input: 1.80,
-      output: 7.20,
-      cached: 0.90,
-      reasoning: 10.80,
-      cache_creation: 1.80
-    },
-    "kimi-latest": {
-      input: 1.00,
-      output: 4.00,
-      cached: 0.50,
-      reasoning: 6.00,
-      cache_creation: 1.00
-    }
-  },
-
-  // MiniMax
-  minimax: {
-    "MiniMax-M2.7": {
-      input: 0.50,
-      output: 2.00,
-      cached: 0.25,
-      reasoning: 3.00,
-      cache_creation: 0.50
-    },
-    "MiniMax-M2.5": {
-      input: 0.50,
-      output: 2.00,
-      cached: 0.25,
-      reasoning: 3.00,
-      cache_creation: 0.50
-    },
-    "MiniMax-M2.1": {
-      input: 0.50,
-      output: 2.00,
-      cached: 0.25,
-      reasoning: 3.00,
-      cache_creation: 0.50
-    }
-  },
-
-  // DeepSeek (official API pricing: https://api-docs.deepseek.com/quick_start/pricing)
-  deepseek: {
-    "deepseek-chat": {
-      input: 0.28,
-      output: 0.42,
-      cached: 0.028,
-      reasoning: 0.42,
-      cache_creation: 0.28
-    },
-    "deepseek-reasoner": {
-      input: 0.28,
-      output: 0.42,
-      cached: 0.028,
-      reasoning: 0.42,
-      cache_creation: 0.28
-    }
-  }
-};
+// Pricing rates for AI models — all rates in $/1M tokens
+//
+// Fallback order (first match wins):
+//   1. PROVIDER_PRICING[provider][model]  — provider-specific override
+//   2. MODEL_PRICING[model]               — canonical model price (provider-agnostic)
+//   3. PATTERN_PRICING                    — glob pattern match (e.g. "codex-*")
 
 /**
- * Provider-agnostic fallback pricing, keyed by model base name.
- * Used when provider-specific lookup fails (e.g. openrouter, cu, fireworks use prefixed model IDs).
- * Strip prefix before lookup: "deepseek/deepseek-chat" → "deepseek-chat"
+ * Canonical model pricing — provider-agnostic.
+ * Cover all known models; deduplicated across providers.
  */
 export const MODEL_PRICING = {
-  // DeepSeek
-  "deepseek-chat":     { input: 0.28, output: 0.42, cached: 0.028, reasoning: 0.42, cache_creation: 0.28 },
-  "deepseek-reasoner": { input: 0.28, output: 0.42, cached: 0.028, reasoning: 0.42, cache_creation: 0.28 },
+  // === Anthropic / Claude ===
+  "claude-opus-4-6":              { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 25.00,  cache_creation: 6.25  },
+  "claude-opus-4-5-20251101":     { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 25.00,  cache_creation: 6.25  },
+  "claude-sonnet-4-6":            { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 15.00,  cache_creation: 3.75  },
+  "claude-sonnet-4-5-20250929":   { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 15.00,  cache_creation: 3.75  },
+  "claude-haiku-4-5-20251001":    { input: 1.00,  output: 5.00,  cached: 0.10,  reasoning: 5.00,   cache_creation: 1.25  },
+  "claude-sonnet-4-20250514":     { input: 3.00,  output: 15.00, cached: 1.50,  reasoning: 15.00,  cache_creation: 3.00  },
+  "claude-opus-4-20250514":       { input: 15.00, output: 25.00, cached: 7.50,  reasoning: 112.50, cache_creation: 15.00 },
+  "claude-3-5-sonnet-20241022":   { input: 3.00,  output: 15.00, cached: 1.50,  reasoning: 15.00,  cache_creation: 3.00  },
+  "claude-haiku-4.5":             { input: 0.50,  output: 2.50,  cached: 0.05,  reasoning: 3.75,   cache_creation: 0.50  },
+  "claude-opus-4.1":              { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 37.50,  cache_creation: 5.00  },
+  "claude-opus-4.5":              { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 37.50,  cache_creation: 5.00  },
+  "claude-opus-4.6":              { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 37.50,  cache_creation: 5.00  },
+  "claude-sonnet-4":              { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 22.50,  cache_creation: 3.00  },
+  "claude-sonnet-4.5":            { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 22.50,  cache_creation: 3.00  },
+  "claude-sonnet-4.6":            { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 22.50,  cache_creation: 3.00  },
+  "claude-opus-4-5-thinking":     { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 37.50,  cache_creation: 5.00  },
+  "claude-opus-4-6-thinking":     { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 37.50,  cache_creation: 5.00  },
+
+  // === OpenAI / GPT ===
+  "gpt-3.5-turbo":                { input: 0.50,  output: 1.50,  cached: 0.25,  reasoning: 2.25,   cache_creation: 0.50  },
+  "gpt-4":                        { input: 2.50,  output: 10.00, cached: 1.25,  reasoning: 15.00,  cache_creation: 2.50  },
+  "gpt-4-turbo":                  { input: 10.00, output: 30.00, cached: 5.00,  reasoning: 45.00,  cache_creation: 10.00 },
+  "gpt-4o":                       { input: 2.50,  output: 10.00, cached: 1.25,  reasoning: 15.00,  cache_creation: 2.50  },
+  "gpt-4o-mini":                  { input: 0.15,  output: 0.60,  cached: 0.075, reasoning: 0.90,   cache_creation: 0.15  },
+  "gpt-4.1":                      { input: 2.50,  output: 10.00, cached: 1.25,  reasoning: 15.00,  cache_creation: 2.50  },
+  "gpt-5":                        { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  },
+  "gpt-5-mini":                   { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  },
+  "gpt-5-codex":                  { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  },
+  "gpt-5.1":                      { input: 4.00,  output: 16.00, cached: 2.00,  reasoning: 24.00,  cache_creation: 4.00  },
+  "gpt-5.1-codex":                { input: 4.00,  output: 16.00, cached: 2.00,  reasoning: 24.00,  cache_creation: 4.00  },
+  "gpt-5.1-codex-mini":           { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  },
+  "gpt-5.1-codex-mini-high":      { input: 2.00,  output: 8.00,  cached: 1.00,  reasoning: 12.00,  cache_creation: 2.00  },
+  "gpt-5.1-codex-max":            { input: 8.00,  output: 32.00, cached: 4.00,  reasoning: 48.00,  cache_creation: 8.00  },
+  "gpt-5.2":                      { input: 5.00,  output: 20.00, cached: 2.50,  reasoning: 30.00,  cache_creation: 5.00  },
+  "gpt-5.2-codex":                { input: 5.00,  output: 20.00, cached: 2.50,  reasoning: 30.00,  cache_creation: 5.00  },
+  "gpt-5.3-codex":                { input: 6.00,  output: 24.00, cached: 3.00,  reasoning: 36.00,  cache_creation: 6.00  },
+  "gpt-5.3-codex-xhigh":         { input: 10.00, output: 40.00, cached: 5.00,  reasoning: 60.00,  cache_creation: 10.00 },
+  "gpt-5.3-codex-high":          { input: 8.00,  output: 32.00, cached: 4.00,  reasoning: 48.00,  cache_creation: 8.00  },
+  "gpt-5.3-codex-low":           { input: 4.00,  output: 16.00, cached: 2.00,  reasoning: 24.00,  cache_creation: 4.00  },
+  "gpt-5.3-codex-none":          { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  },
+  "gpt-5.3-codex-spark":         { input: 3.00,  output: 12.00, cached: 0.30,  reasoning: 12.00,  cache_creation: 3.00  },
+  "o1":                           { input: 15.00, output: 60.00, cached: 7.50,  reasoning: 90.00,  cache_creation: 15.00 },
+  "o1-mini":                      { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  },
+
+  // === Gemini ===
+  "gemini-3-flash-preview":       { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  },
+  "gemini-3-pro-preview":         { input: 2.00,  output: 12.00, cached: 0.25,  reasoning: 18.00,  cache_creation: 2.00  },
+  "gemini-3.1-pro-low":           { input: 2.00,  output: 12.00, cached: 0.25,  reasoning: 18.00,  cache_creation: 2.00  },
+  "gemini-3.1-pro-high":          { input: 4.00,  output: 18.00, cached: 0.50,  reasoning: 27.00,  cache_creation: 4.00  },
+  "gemini-3-flash":               { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  },
+  "gemini-2.5-pro":               { input: 2.00,  output: 12.00, cached: 0.25,  reasoning: 18.00,  cache_creation: 2.00  },
+  "gemini-2.5-flash":             { input: 0.30,  output: 2.50,  cached: 0.03,  reasoning: 3.75,   cache_creation: 0.30  },
+  "gemini-2.5-flash-lite":        { input: 0.15,  output: 1.25,  cached: 0.015, reasoning: 1.875,  cache_creation: 0.15  },
+
+  // === Qwen ===
+  "qwen3-coder-plus":             { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  },
+  "qwen3-coder-flash":            { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+
+  // === Kimi ===
+  "kimi-k2":                      { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  },
+  "kimi-k2-thinking":             { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  },
+  "kimi-k2.5":                    { input: 1.20,  output: 4.80,  cached: 0.60,  reasoning: 7.20,   cache_creation: 1.20  },
+  "kimi-k2.5-thinking":           { input: 1.80,  output: 7.20,  cached: 0.90,  reasoning: 10.80,  cache_creation: 1.80  },
+  "kimi-latest":                  { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  },
+
+  // === DeepSeek ===
+  "deepseek-chat":                { input: 0.28,  output: 0.42,  cached: 0.028, reasoning: 0.42,   cache_creation: 0.28  },
+  "deepseek-reasoner":            { input: 0.28,  output: 0.42,  cached: 0.028, reasoning: 0.42,   cache_creation: 0.28  },
+  "deepseek-r1":                  { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  },
+  "deepseek-v3.2-chat":           { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "deepseek-v3.2-reasoner":       { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  },
+
+  // === GLM ===
+  "glm-4.6":                      { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "glm-4.6v":                     { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  },
+  "glm-4.7":                      { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  },
+  "glm-5":                        { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  },
+
+  // === MiniMax ===
+  "MiniMax-M2.1":                 { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "MiniMax-M2.5":                 { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "MiniMax-M2.7":                 { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "minimax-m2.1":                 { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "minimax-m2.5":                 { input: 0.60,  output: 2.40,  cached: 0.30,  reasoning: 3.60,   cache_creation: 0.60  },
+
+  // === Grok ===
+  "grok-code-fast-1":             { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+
+  // === OpenRouter fallback ===
+  "auto":                         { input: 2.00,  output: 8.00,  cached: 1.00,  reasoning: 12.00,  cache_creation: 2.00  },
+
+  // === Misc ===
+  "oswe-vscode-prime":            { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  },
+  "gpt-oss-120b-medium":          { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "vision-model":                 { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  },
+  "coder-model":                  { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  },
 };
 
 /**
- * Get pricing for a specific provider and model
- * @param {string} provider - Provider ID (e.g., "openai", "cc", "gc")
- * @param {string} model - Model ID
- * @returns {object|null} Pricing object or null if not found
+ * Provider-specific pricing overrides.
+ * Only include entries where price DIFFERS from MODEL_PRICING.
+ * Keyed by provider alias (cc, cx, gc, gh, ...) or provider id (openai, anthropic, ...).
  */
-export function getPricingForModel(provider, model) {
-  if (!provider || !model) return null;
+export const PROVIDER_PRICING = {
+  // GitHub Copilot (gh) — gpt-5.3-codex has different rate than canonical
+  gh: {
+    "gpt-5.3-codex": { input: 1.75, output: 14.00, cached: 0.175, reasoning: 14.00, cache_creation: 1.75 },
+  },
+};
 
-  const providerPricing = DEFAULT_PRICING[provider];
-  if (!providerPricing) return null;
+/**
+ * Pattern-based pricing fallback — matched when no exact model entry found.
+ * Patterns use simple glob: "*" matches any substring.
+ * First match wins — order matters.
+ */
+export const PATTERN_PRICING = [
+  // --- Codex variants ---
+  { pattern: "*-codex-xhigh",   pricing: { input: 10.00, output: 40.00, cached: 5.00,  reasoning: 60.00,  cache_creation: 10.00 } },
+  { pattern: "*-codex-high",    pricing: { input: 8.00,  output: 32.00, cached: 4.00,  reasoning: 48.00,  cache_creation: 8.00  } },
+  { pattern: "*-codex-max",     pricing: { input: 8.00,  output: 32.00, cached: 4.00,  reasoning: 48.00,  cache_creation: 8.00  } },
+  { pattern: "*-codex-mini-*",  pricing: { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  } },
+  { pattern: "*-codex-mini",    pricing: { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  } },
+  { pattern: "*-codex-low",     pricing: { input: 4.00,  output: 16.00, cached: 2.00,  reasoning: 24.00,  cache_creation: 4.00  } },
+  { pattern: "*-codex-none",    pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+  { pattern: "*-codex-spark",   pricing: { input: 3.00,  output: 12.00, cached: 0.30,  reasoning: 12.00,  cache_creation: 3.00  } },
+  { pattern: "codex-*",         pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+  { pattern: "*-codex",         pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
 
-  return providerPricing[model] || null;
+  // --- Claude ---
+  { pattern: "claude-opus-*",   pricing: { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 25.00,  cache_creation: 6.25  } },
+  { pattern: "claude-sonnet-*", pricing: { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 15.00,  cache_creation: 3.75  } },
+  { pattern: "claude-haiku-*",  pricing: { input: 1.00,  output: 5.00,  cached: 0.10,  reasoning: 5.00,   cache_creation: 1.25  } },
+  { pattern: "claude-*",        pricing: { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 15.00,  cache_creation: 3.75  } },
+
+  // --- Gemini (specific trước, chung sau) ---
+  { pattern: "gemini-*-flash-lite", pricing: { input: 0.15, output: 1.25, cached: 0.015, reasoning: 1.875, cache_creation: 0.15 } },
+  { pattern: "gemini-*-flash",  pricing: { input: 0.30,  output: 2.50,  cached: 0.03,  reasoning: 3.75,   cache_creation: 0.30  } },
+  { pattern: "gemini-*-pro",    pricing: { input: 2.00,  output: 12.00, cached: 0.25,  reasoning: 18.00,  cache_creation: 2.00  } },
+  { pattern: "gemini-3-*",      pricing: { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  } },
+  { pattern: "gemini-2.5-*",    pricing: { input: 0.30,  output: 2.50,  cached: 0.03,  reasoning: 3.75,   cache_creation: 0.30  } },
+  { pattern: "gemini-*",        pricing: { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  } },
+
+  // --- GPT (specific trước, chung sau) ---
+  { pattern: "gpt-5.3-*",       pricing: { input: 6.00,  output: 24.00, cached: 3.00,  reasoning: 36.00,  cache_creation: 6.00  } },
+  { pattern: "gpt-5.2-*",       pricing: { input: 5.00,  output: 20.00, cached: 2.50,  reasoning: 30.00,  cache_creation: 5.00  } },
+  { pattern: "gpt-5.1-*",       pricing: { input: 4.00,  output: 16.00, cached: 2.00,  reasoning: 24.00,  cache_creation: 4.00  } },
+  { pattern: "gpt-5-*",         pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+  { pattern: "gpt-5*",          pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+  { pattern: "gpt-4o-*",        pricing: { input: 0.15,  output: 0.60,  cached: 0.075, reasoning: 0.90,   cache_creation: 0.15  } },
+  { pattern: "gpt-4o",          pricing: { input: 2.50,  output: 10.00, cached: 1.25,  reasoning: 15.00,  cache_creation: 2.50  } },
+  { pattern: "gpt-4*",          pricing: { input: 2.50,  output: 10.00, cached: 1.25,  reasoning: 15.00,  cache_creation: 2.50  } },
+
+  // --- o1 / o-series ---
+  { pattern: "o1-*",            pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+  { pattern: "o1",              pricing: { input: 15.00, output: 60.00, cached: 7.50,  reasoning: 90.00,  cache_creation: 15.00 } },
+  { pattern: "o3-*",            pricing: { input: 10.00, output: 40.00, cached: 5.00,  reasoning: 60.00,  cache_creation: 10.00 } },
+  { pattern: "o4-*",            pricing: { input: 2.00,  output: 8.00,  cached: 1.00,  reasoning: 12.00,  cache_creation: 2.00  } },
+
+  // --- Qwen ---
+  { pattern: "qwen3-coder-*",   pricing: { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  } },
+  { pattern: "qwen*-coder-*",   pricing: { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  } },
+  { pattern: "qwen*",           pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+
+  // --- Kimi ---
+  { pattern: "kimi-*-thinking",  pricing: { input: 1.80,  output: 7.20,  cached: 0.90,  reasoning: 10.80,  cache_creation: 1.80  } },
+  { pattern: "kimi-k2*",        pricing: { input: 1.20,  output: 4.80,  cached: 0.60,  reasoning: 7.20,   cache_creation: 1.20  } },
+  { pattern: "kimi-*",          pricing: { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  } },
+
+  // --- DeepSeek ---
+  { pattern: "deepseek-*reasoner*", pricing: { input: 0.75, output: 3.00, cached: 0.375, reasoning: 4.50, cache_creation: 0.75 } },
+  { pattern: "deepseek-r*",     pricing: { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  } },
+  { pattern: "deepseek-v*",     pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+  { pattern: "deepseek-*",      pricing: { input: 0.28,  output: 0.42,  cached: 0.028, reasoning: 0.42,   cache_creation: 0.28  } },
+
+  // --- GLM ---
+  { pattern: "glm-5*",          pricing: { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  } },
+  { pattern: "glm-4*",          pricing: { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  } },
+  { pattern: "glm-*",           pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+
+  // --- MiniMax ---
+  { pattern: "MiniMax-*",       pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+  { pattern: "minimax-*",       pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+
+  // --- Grok ---
+  { pattern: "grok-code-*",     pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+  { pattern: "grok-*",          pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+];
+
+/**
+ * Match a model ID against a glob pattern (* = wildcard).
+ */
+function matchPattern(pattern, model) {
+  const regex = new RegExp("^" + pattern.split("*").map(s => s.replace(/[.*+?^${}()|[\]\\]/g, "\\$&")).join(".*") + "$");
+  return regex.test(model);
 }
 
 /**
- * Get all pricing data
- * @returns {object} All default pricing
+ * Resolve pricing for a model using the 3-step fallback chain:
+ *   1. PROVIDER_PRICING[provider][model]
+ *   2. MODEL_PRICING[model]
+ *   3. PATTERN_PRICING (glob match)
+ *
+ * @param {string} provider
+ * @param {string} model
+ * @returns {object|null}
+ */
+export function getPricingForModel(provider, model) {
+  if (!model) return null;
+
+  // 1. Provider-specific override
+  if (provider && PROVIDER_PRICING[provider]?.[model]) {
+    return PROVIDER_PRICING[provider][model];
+  }
+
+  // 2. Canonical model pricing (strip vendor prefix if needed: "deepseek/deepseek-chat" → "deepseek-chat")
+  const baseModel = model.includes("/") ? model.split("/").pop() : model;
+  if (MODEL_PRICING[baseModel]) return MODEL_PRICING[baseModel];
+  if (MODEL_PRICING[model]) return MODEL_PRICING[model];
+
+  // 3. Pattern match
+  for (const { pattern, pricing } of PATTERN_PRICING) {
+    if (matchPattern(pattern, baseModel) || matchPattern(pattern, model)) {
+      return pricing;
+    }
+  }
+
+  return null;
+}
+
+/**
+ * Get all provider pricing (for UI / API).
+ * Returns PROVIDER_PRICING — consumers should fall back to MODEL_PRICING for unlisted models.
  */
 export function getDefaultPricing() {
-  return DEFAULT_PRICING;
+  return PROVIDER_PRICING;
 }
 
 /**
  * Format cost for display
- * @param {number} cost - Cost in dollars
- * @returns {string} Formatted cost string
+ * @param {number} cost
+ * @returns {string}
  */
 export function formatCost(cost) {
   if (cost === null || cost === undefined || isNaN(cost)) return "$0.00";
@@ -832,44 +260,36 @@ export function formatCost(cost) {
 
 /**
  * Calculate cost from tokens and pricing
- * @param {object} tokens - Token counts
- * @param {object} pricing - Pricing object
- * @returns {number} Cost in dollars
+ * @param {object} tokens
+ * @param {object} pricing
+ * @returns {number} cost in dollars
  */
 export function calculateCostFromTokens(tokens, pricing) {
   if (!tokens || !pricing) return 0;
 
   let cost = 0;
 
-  // Input tokens (non-cached)
   const inputTokens = tokens.prompt_tokens || tokens.input_tokens || 0;
   const cachedTokens = tokens.cached_tokens || tokens.cache_read_input_tokens || 0;
   const nonCachedInput = Math.max(0, inputTokens - cachedTokens);
 
-  cost += (nonCachedInput * (pricing.input / 1000000));
+  cost += nonCachedInput * (pricing.input / 1000000);
 
-  // Cached tokens
   if (cachedTokens > 0) {
-    const cachedRate = pricing.cached || pricing.input; // Fallback to input rate
-    cost += (cachedTokens * (cachedRate / 1000000));
+    cost += cachedTokens * ((pricing.cached || pricing.input) / 1000000);
   }
 
-  // Output tokens
   const outputTokens = tokens.completion_tokens || tokens.output_tokens || 0;
-  cost += (outputTokens * (pricing.output / 1000000));
+  cost += outputTokens * (pricing.output / 1000000);
 
-  // Reasoning tokens
   const reasoningTokens = tokens.reasoning_tokens || 0;
   if (reasoningTokens > 0) {
-    const reasoningRate = pricing.reasoning || pricing.output; // Fallback to output rate
-    cost += (reasoningTokens * (reasoningRate / 1000000));
+    cost += reasoningTokens * ((pricing.reasoning || pricing.output) / 1000000);
   }
 
-  // Cache creation tokens
   const cacheCreationTokens = tokens.cache_creation_input_tokens || 0;
   if (cacheCreationTokens > 0) {
-    const cacheCreationRate = pricing.cache_creation || pricing.input; // Fallback to input rate
-    cost += (cacheCreationTokens * (cacheCreationRate / 1000000));
+    cost += cacheCreationTokens * ((pricing.cache_creation || pricing.input) / 1000000);
   }
 
   return cost;