merge: Dev_20260801 → main（v0.6.9：仪表盘缓存命中列 + 人民币计价 + 性能优化）

package.json

@@ -1,7 +1,7 @@
 {
   "name": "kimiswitch",
   "private": true,
-  "version": "0.6.8",
+  "version": "0.6.9",
   "type": "module",
   "scripts": {
     "dev": "vite",

src-tauri/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kimiswitch"
-version = "0.6.8"
+version = "0.6.9"
 description = "Kimi Switch - model config manager"
 authors = ["you"]
 edition = "2021"

src-tauri/src/dashboard.rs

@@ -610,7 +610,26 @@ fn strip_context_suffix(model_name: &str) -> Option<String> {
     })
 }
 
+/// Memoized price resolution. `match_price` is called once per usage record
+/// during the scan, and the models.dev suffix scan is O(5910) per miss — with
+/// thousands of records sharing a handful of model names, memoizing the result
+/// turns the whole scan from O(records × 5910) into O(unique_models × 5910).
+static MATCH_PRICE_MEMO: OnceLock<Mutex<HashMap<String, (String, f64, f64, f64, bool)>>> =
+    OnceLock::new();
+
+/// Resolve a model's price, caching per-model results (hits and misses).
 fn match_price(model_name: &str) -> (String, f64, f64, f64, bool) {
+    let key = model_name.to_ascii_lowercase();
+    let memo = MATCH_PRICE_MEMO.get_or_init(|| Mutex::new(HashMap::new()));
+    if let Some(hit) = memo.lock().unwrap().get(&key) {
+        return hit.clone();
+    }
+    let result = match_price_inner(model_name);
+    memo.lock().unwrap().insert(key, result.clone());
+    result
+}
+
+fn match_price_inner(model_name: &str) -> (String, f64, f64, f64, bool) {
     let bare = match model_name.rsplit_once('/') {
         Some((_, b)) => b,
         None => model_name,
@@ -1537,6 +1556,7 @@ pub fn get_prices() -> PricesResult {
 
 #[tauri::command]
 pub fn get_summary(home_override: Option<String>, range: Option<String>, refresh: Option<bool>) -> SummaryResult {
+    let t0 = std::time::Instant::now();
     let refresh = refresh.unwrap_or(false);
     let home = resolve_kimi_home(home_override);
     let now_ms = std::time::SystemTime::now()
@@ -1544,6 +1564,7 @@ pub fn get_summary(home_override: Option<String>, range: Option<String>, refresh
     let r = range.unwrap_or_else(|| "30d".into());
 
     let (records, meta) = scan_usage_cached(&home, refresh);
+    let t1 = std::time::Instant::now();
     let stats = aggregate(&records, &r, now_ms);
     let all_stats = aggregate(&records, "all", now_ms);
     let heatmap = build_heatmap(&records, now_ms);
@@ -1568,6 +1589,15 @@ pub fn get_summary(home_override: Option<String>, range: Option<String>, refresh
         name, provider: std::env::var("KIMI_MODEL_PROVIDER").ok(), model: std::env::var("KIMI_MODEL_ID").ok(),
     });
 
+    let t2 = std::time::Instant::now();
+    let scan_ms = t1.duration_since(t0).as_secs_f64() * 1000.0;
+    let aggregate_ms = t2.duration_since(t1).as_secs_f64() * 1000.0;
+    let total_ms = t2.duration_since(t0).as_secs_f64() * 1000.0;
+    eprintln!(
+        "[dashboard] get_summary range={} records={} scan={:.0}ms aggregate={:.0}ms total={:.0}ms",
+        r, records.len(), scan_ms, aggregate_ms, total_ms
+    );
+
     SummaryResult {
         home: home.to_string_lossy().to_string(),
         valid: is_kimi_home(&home),

src-tauri/tauri.conf.json

@@ -1,6 +1,6 @@
 {
   "productName": "Kimi Switch",
-  "version": "0.6.8",
+  "version": "0.6.9",
   "identifier": "com.kimiswitch.app",
   "build": {
     "beforeDevCommand": "npm run dev",

src/components/dashboard/DailyBars.tsx

@@ -1,7 +1,8 @@
 import { useMemo, useState } from "react";
 import { createPortal } from "react-dom";
 import { useTranslation } from "../../i18n";
-import { fmtPct, fmtTokens, fmtUsd } from "../../lib/dashboard-format";
+import { useCurrency } from "../../hooks/useCurrency";
+import { fmtPct, fmtTokens } from "../../lib/dashboard-format";
 import type { DailyRow } from "../../types/dashboard";
 import { DailyDetailModal } from "./DailyDetailModal";
 
@@ -43,6 +44,7 @@ function shortModel(model: string): string {
 
 export function DailyBars({ daily, dimension, names, unknownProviderLabel }: DailyBarsProps) {
   const { t } = useTranslation();
+  const { money } = useCurrency();
   const [selected, setSelected] = useState<DailyRow | null>(null);
 
   const colorMap = useMemo(() => {
@@ -103,7 +105,7 @@ export function DailyBars({ daily, dimension, names, unknownProviderLabel }: Dai
 
           const tooltipLines = [
             d.date,
-            `${t("totalTokens")}: ${fmtTokens(d.totalTokens)} · ${fmtUsd(d.costUsd)} · ${t("colCacheHit")} ${fmtPct(d.cacheHitRate)}`,
+            `${t("totalTokens")}: ${fmtTokens(d.totalTokens)} · ${money(d.costUsd)} · ${t("colCacheHit")} ${fmtPct(d.cacheHitRate)}`,
             ...segments.map((s) => `${displayName(s.key)}: ${fmtTokens(s.tokens)}`),
             "",
             t("doubleClickHint"),

src/components/dashboard/DailyDetailModal.tsx

@@ -1,6 +1,7 @@
 import { useEffect } from "react";
 import { useTranslation } from "../../i18n";
-import { fmtPct, fmtTokens, fmtUsd } from "../../lib/dashboard-format";
+import { useCurrency } from "../../hooks/useCurrency";
+import { fmtPct, fmtTokens } from "../../lib/dashboard-format";
 import type { DailyRow } from "../../types/dashboard";
 import { modelColor } from "./DailyBars";
 
@@ -22,6 +23,7 @@ function shortModel(model: string): string {
 
 export function DailyDetailModal({ day, dimension, colorMap, unknownProviderLabel, onClose }: DailyDetailModalProps) {
   const { t } = useTranslation();
+  const { money } = useCurrency();
 
   useEffect(() => {
     const onKey = (e: KeyboardEvent) => {
@@ -106,7 +108,7 @@ export function DailyDetailModal({ day, dimension, colorMap, unknownProviderLabe
         {/* Day summary chips */}
         <div className="grid grid-cols-2 gap-2 px-5 py-3 sm:grid-cols-4">
           <SummaryChip label={t("totalTokens")} value={fmtTokens(day.totalTokens)} tone="default" />
-          <SummaryChip label={t("cost")} value={fmtUsd(day.costUsd)} tone="orange" />
+          <SummaryChip label={t("cost")} value={money(day.costUsd)} tone="orange" />
           <SummaryChip label={t("requests")} value={Math.round(day.requests).toLocaleString()} tone="blue" />
           <SummaryChip label={t("cacheHitRate")} value={fmtPct(day.cacheHitRate)} tone="green" />
         </div>

src/components/dashboard/DashboardPage.tsx

@@ -1,13 +1,20 @@
 import { useState, type ReactNode } from "react";
 import { useDashboard, type DashboardRange } from "../../hooks/useDashboard";
+import { useCurrency } from "../../hooks/useCurrency";
 import { useTranslation } from "../../i18n";
-import { fmtInt, fmtPct, fmtTime, fmtTokens, fmtUsd } from "../../lib/dashboard-format";
+import { fmtInt, fmtPct, fmtTime, fmtTokens } from "../../lib/dashboard-format";
 import { DailyBars } from "./DailyBars";
 import { Heatmap } from "./Heatmap";
 import { TrendLineChart } from "./TrendLineChart";
 
 const RANGES: DashboardRange[] = ["today", "7d", "30d", "all"];
 
+/** Per-request cache hit rate: cached tokens / total input tokens. */
+function cacheHitOf(r: { inputOther: number; inputCacheRead: number }): number {
+  const denom = r.inputOther + r.inputCacheRead;
+  return denom > 0 ? r.inputCacheRead / denom : 0;
+}
+
 function Card({
   title,
   subtitle,
@@ -44,7 +51,8 @@ function KpiCard({ label, value, sub }: { label: string; value: string; sub?: st
 
 export function DashboardPage() {
   const { t } = useTranslation();
-  const { range, changeRange, data, loading, error, refresh } = useDashboard();
+  const { money } = useCurrency();
+  const { range, changeRange, data, loading, error, refresh, loadStats } = useDashboard();
   const [showAllModels, setShowAllModels] = useState(false);
   const [recentPage, setRecentPage] = useState(1);
   const [trendTab, setTrendTab] = useState<"daily" | "model" | "provider">("daily");
@@ -120,28 +128,35 @@ export function DashboardPage() {
     { label: t("kpiCacheCreation"), value: fmtTokens(totals.inputCacheCreation), sub: fmtInt(totals.inputCacheCreation) },
     { label: t("kpiCacheHit"), value: fmtPct(totals.cacheHitRate) },
     { label: t("totalTokens"), value: fmtTokens(totals.totalTokens), sub: fmtInt(totals.totalTokens) },
-    { label: t("kpiCost"), value: fmtUsd(totals.costUsd) },
+    { label: t("kpiCost"), value: money(totals.costUsd) },
   ];
 
   return (
     <div className="h-full overflow-auto p-4 space-y-4">
       {/* Range tabs + refresh */}
       <div className="flex items-center justify-between">
-        <div className="flex items-center bg-input border border-border rounded p-0.5">
+        <div className="flex items-center gap-3">
-          {RANGES.map((r) => (
+          <div className="flex items-center bg-input border border-border rounded p-0.5">
-            <button
+            {RANGES.map((r) => (
-              key={r}
+              <button
-              type="button"
+                key={r}
-              onClick={() => changeRange(r)}
+                type="button"
-              className={`px-3 py-1 text-sm rounded transition-colors ${
+                onClick={() => changeRange(r)}
-                range === r
+                className={`px-3 py-1 text-sm rounded transition-colors ${
-                  ? "bg-blue-600 text-white"
+                  range === r
-                  : "text-content-muted hover:text-content-primary"
+                    ? "bg-blue-600 text-white"
-              }`}
+                    : "text-content-muted hover:text-content-primary"
-            >
+                }`}
-              {RANGE_LABELS[r]}
+              >
-            </button>
+                {RANGE_LABELS[r]}
-          ))}
+              </button>
+            ))}
+          </div>
+          {loadStats && !loading && (
+            <span className="hidden text-[10px] text-content-muted sm:inline">
+              {t("loadStats", { ms: loadStats.ms, kb: loadStats.kb })}
+            </span>
+          )}
         </div>
         <button
           type="button"
@@ -175,7 +190,7 @@ export function DashboardPage() {
                 {fmtTokens(tot.totalTokens)}
               </div>
               <div className="mt-1 text-[11px] text-content-muted">
-                {fmtInt(tot.requests)} {t("unitTimes")} · {fmtUsd(tot.costUsd)} · {fmtPct(tot.cacheHitRate)}
+                {fmtInt(tot.requests)} {t("unitTimes")} · {money(tot.costUsd)} · {fmtPct(tot.cacheHitRate)}
               </div>
             </button>
           );
@@ -305,7 +320,7 @@ export function DashboardPage() {
                       <td className="py-2 pr-3 text-right text-content-muted">{fmtInt(m.requests)}</td>
                       <td className="py-2 pr-3 text-right text-content-muted">{fmtTokens(m.totalTokens)}</td>
                       <td className="py-2 pr-3 text-right text-content-muted">{fmtPct(m.cacheHitRate)}</td>
-                      <td className="py-2 text-right text-orange-600 dark:text-orange-400">{fmtUsd(m.costUsd)}</td>
+                      <td className="py-2 text-right text-orange-600 dark:text-orange-400">{money(m.costUsd)}</td>
                     </tr>
                   ))}
                 </tbody>
@@ -346,6 +361,7 @@ export function DashboardPage() {
                     <th className="pb-2 pr-4 font-normal text-right">{t("colInput")}</th>
                     <th className="pb-2 pr-4 font-normal text-right">{t("colOutput")}</th>
                     <th className="pb-2 pr-4 font-normal text-right">{t("colCacheRead")}</th>
+                    <th className="pb-2 pr-4 font-normal text-right">{t("colCacheHit")}</th>
                     <th className="pb-2 font-normal text-right">{t("cost")}</th>
                   </tr>
                 </thead>
@@ -359,7 +375,8 @@ export function DashboardPage() {
                       <td className="py-1.5 pr-4 text-right text-content-muted">{fmtTokens(r.inputOther)}</td>
                       <td className="py-1.5 pr-4 text-right text-content-muted">{fmtTokens(r.output)}</td>
                       <td className="py-1.5 pr-4 text-right text-content-muted">{fmtTokens(r.inputCacheRead)}</td>
-                      <td className="py-1.5 text-right text-orange-600 dark:text-orange-400">{fmtUsd(r.costUsd)}</td>
+                      <td className="py-1.5 pr-4 text-right text-content-muted">{fmtPct(cacheHitOf(r))}</td>
+                      <td className="py-1.5 text-right text-orange-600 dark:text-orange-400">{money(r.costUsd)}</td>
                     </tr>
                   ))}
                 </tbody>

src/components/dashboard/Heatmap.tsx

@@ -1,7 +1,8 @@
 import { useEffect, useMemo, useRef, useState, type MouseEvent } from "react";
 import { createPortal } from "react-dom";
 import { useTranslation } from "../../i18n";
-import { fmtPct, fmtTokens, fmtUsd } from "../../lib/dashboard-format";
+import { useCurrency } from "../../hooks/useCurrency";
+import { fmtPct, fmtTokens } from "../../lib/dashboard-format";
 import type { HeatmapCell, HeatmapData } from "../../types/dashboard";
 
 interface HeatmapProps {
@@ -220,6 +221,7 @@ export function Heatmap({ heatmap }: HeatmapProps) {
 
 function HeatmapTooltip({ state }: { state: HoverState }) {
   const { t } = useTranslation();
+  const { money } = useCurrency();
   const { cell, x, y } = state;
   // Position above the cursor with a small offset; flip below if too close to top.
   const offset = 14;
@@ -239,7 +241,7 @@ function HeatmapTooltip({ state }: { state: HoverState }) {
     >
       <div className="font-medium text-content-primary">{cell.date}</div>
       <div className="text-content-muted">
-        {fmtTokens(cell.totalTokens)} · {fmtUsd(cell.costUsd)} ·{" "}
+        {fmtTokens(cell.totalTokens)} · {money(cell.costUsd)} ·{" "}
         {fmtInt(cell.requests)} {t("requests")}
       </div>
       <div className="text-content-muted">

src/hooks/useCurrency.ts

@@ -0,0 +1,83 @@
+import { useCallback, useEffect, useState } from "react";
+import { useTranslation } from "../i18n";
+import { fmtUsd } from "../lib/dashboard-format";
+
+/**
+ * USD → CNY exchange rate for the dashboard.
+ * Chinese mode shows prices in CNY (latest available rate), English stays in
+ * USD. The rate is fetched once and cached in localStorage for 12h; any
+ * failure falls back to a conservative fixed rate so the UI never blocks.
+ */
+
+const FALLBACK_RATE = 7.2;
+const CACHE_KEY = "kimi-switch-cny-rate";
+const CACHE_TTL = 12 * 60 * 60 * 1000; // 12 hours
+
+function readCachedRate(): number | null {
+  try {
+    const raw = localStorage.getItem(CACHE_KEY);
+    if (!raw) return null;
+    const { ts, rate } = JSON.parse(raw) as { ts: number; rate: number };
+    if (Date.now() - ts > CACHE_TTL || !(rate > 1)) return null;
+    return rate;
+  } catch {
+    return null;
+  }
+}
+
+function writeCachedRate(rate: number) {
+  try {
+    localStorage.setItem(CACHE_KEY, JSON.stringify({ ts: Date.now(), rate }));
+  } catch {
+    /* ignore */
+  }
+}
+
+export function useCurrency() {
+  const { lang } = useTranslation();
+  const [rate, setRate] = useState<number | null>(null);
+
+  useEffect(() => {
+    let cancelled = false;
+    (async () => {
+      const cached = readCachedRate();
+      if (cached) {
+        setRate(cached);
+        return;
+      }
+      try {
+        const res = await fetch("https://open.er-api.com/v6/latest/USD");
+        if (res.ok) {
+          const data = (await res.json()) as { rates?: { CNY?: number } };
+          const cny = Number(data.rates?.CNY);
+          if (cny > 1) {
+            if (!cancelled) setRate(cny);
+            writeCachedRate(cny);
+            return;
+          }
+        }
+      } catch {
+        /* network blocked → fallback */
+      }
+      if (!cancelled) setRate(FALLBACK_RATE);
+    })();
+    return () => {
+      cancelled = true;
+    };
+  }, []);
+
+  /** Format a USD cost for display: CNY in Chinese mode, USD otherwise. */
+  const money = useCallback(
+    (usd: number): string => {
+      if (lang !== "zh") return fmtUsd(usd);
+      const v = (Number(usd) || 0) * (rate ?? FALLBACK_RATE);
+      if (v === 0) return "¥0.00";
+      if (v < 0.1) return "¥" + v.toFixed(3);
+      if (v < 1) return "¥" + v.toFixed(2);
+      return "¥" + v.toFixed(2);
+    },
+    [lang, rate],
+  );
+
+  return { money, cnyRate: rate ?? FALLBACK_RATE };
+}

src/hooks/useDashboard.ts

@@ -22,16 +22,23 @@ export function useDashboard() {
   const [data, setData] = useState<SummaryResult | null>(null);
   const [loading, setLoading] = useState(false);
   const [error, setError] = useState<string | null>(null);
+  /** Round-trip timing for the last get_summary call (user-perceived lag). */
+  const [loadStats, setLoadStats] = useState<{ ms: number; kb: number } | null>(null);
 
   const refresh = useCallback(async (force = false) => {
     setLoading(true);
     setError(null);
+    const start = performance.now();
     try {
       const result = await invoke<SummaryResult>("get_summary", {
         range,
         refresh: force,
       });
       setData(result);
+      setLoadStats({
+        ms: Math.round(performance.now() - start),
+        kb: Math.round(JSON.stringify(result).length / 1024),
+      });
     } catch (err) {
       const msg = err instanceof Error ? err.message : String(err);
       setError(msg);
@@ -53,5 +60,5 @@ export function useDashboard() {
     }
   }, []);
 
-  return { range, changeRange, data, loading, error, refresh };
+  return { range, changeRange, data, loading, error, refresh, loadStats };
 }

src/i18n/en.ts

@@ -197,6 +197,7 @@ export const enTranslations: Record<TranslationKey, string> = {
   scanning: "Scanning…",
   refresh: "Refresh",
   refreshing: "Refreshing…",
+  loadStats: "Loaded in {ms} ms · {kb} KB payload",
   refreshed: "Refreshed",
   cancel: "Cancel",
   confirmDeleteTitle: "Delete session forever?",

src/i18n/zh.ts

@@ -194,6 +194,7 @@ export const zhTranslations = {
   scanning: "扫描中…",
   refresh: "刷新",
   refreshing: "刷新中…",
+  loadStats: "加载 {ms} ms · 数据 {kb} KB",
   refreshed: "已刷新",
   cancel: "取消",
   confirmDeleteTitle: "永久删除会话？",

src/lib/models-dev-full.json

@@ -17101,7 +17101,9 @@
           "output_limit": 8192,
           "reasoning": true,
           "tool_call": true,
-          "structured_output": true
+          "structured_output": true,
+          "image": true,
+          "video": true
         },
         {
           "id": "qwen3.6-27b",
@@ -17175,7 +17177,8 @@
           "name": "Abliterated Model Large",
           "cost": {
             "input": 5,
-            "output": 5
+            "output": 5,
+            "cache_read": 0.5
           },
           "context": 1000000,
           "input_limit": 1000000,
@@ -17189,7 +17192,8 @@
           "name": "Abliterated Model",
           "cost": {
             "input": 3,
-            "output": 3
+            "output": 3,
+            "cache_read": 0.3
           },
           "context": 150000,
           "input_limit": 150000,
@@ -18190,7 +18194,9 @@
           "output_limit": 65536,
           "reasoning": true,
           "tool_call": true,
-          "structured_output": true
+          "structured_output": true,
+          "image": true,
+          "video": true
         },
         {
           "id": "gpt-5.2-chat-latest",
@@ -19796,6 +19802,22 @@
           "structured_output": true,
           "image": true
         },
+        {
+          "id": "kimi-k3-fast",
+          "name": "Kimi K3",
+          "cost": {
+            "input": 4.5,
+            "output": 22.5,
+            "cache_read": 0.45
+          },
+          "context": 1040384,
+          "output_limit": 131072,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true,
+          "image": true,
+          "video": true
+        },
         {
           "id": "claude-opus-5",
           "name": "Claude Opus 5",
@@ -29065,7 +29087,9 @@
           "output_limit": 32768,
           "reasoning": true,
           "tool_call": true,
-          "structured_output": true
+          "structured_output": true,
+          "image": true,
+          "video": true
         },
         {
           "id": "Qwen/Qwen3-8B",
@@ -29441,7 +29465,8 @@
           "name": "zai-org/GLM-5",
           "cost": {
             "input": 0.95,
-            "output": 2.55
+            "output": 2.55,
+            "cache_read": 0.2
           },
           "context": 205000,
           "output_limit": 205000,
@@ -29467,6 +29492,7 @@
           "cost": {
             "input": 1.4,
             "output": 4.4,
+            "cache_read": 0.26,
             "cache_write": 0
           },
           "context": 1049000,
@@ -29481,6 +29507,7 @@
           "cost": {
             "input": 1.4,
             "output": 4.4,
+            "cache_read": 0.26,
             "cache_write": 0
           },
           "context": 205000,
@@ -29495,6 +29522,7 @@
           "cost": {
             "input": 1.2,
             "output": 4,
+            "cache_read": 0.24,
             "cache_write": 0
           },
           "context": 200000,
@@ -29824,7 +29852,8 @@
           "name": "MiniMaxAI/MiniMax-M2.5",
           "cost": {
             "input": 0.3,
-            "output": 1.2
+            "output": 1.2,
+            "cache_read": 0.03
           },
           "context": 197000,
           "output_limit": 131000,
@@ -29992,7 +30021,8 @@
           "name": "moonshotai/Kimi-K2.5",
           "cost": {
             "input": 0.45,
-            "output": 2.25
+            "output": 2.25,
+            "cache_read": 0.07
           },
           "context": 262000,
           "output_limit": 262000,
@@ -36790,8 +36820,9 @@
           "id": "zai-org-glm-4.7-flash",
           "name": "GLM 4.7 Flash",
           "cost": {
-            "input": 0.125,
+            "input": 0.06,
-            "output": 0.5
+            "output": 0.4,
+            "cache_read": 0.01
           },
           "context": 128000,
           "output_limit": 16384,
@@ -36825,6 +36856,20 @@
           "structured_output": true,
           "image": true
         },
+        {
+          "id": "deepseek-v4-flash-0731",
+          "name": "DeepSeek V4 Flash 0731",
+          "cost": {
+            "input": 0.175,
+            "output": 0.35,
+            "cache_read": 0.035
+          },
+          "context": 1000000,
+          "output_limit": 32768,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true
+        },
         {
           "id": "qwen3-235b-a22b-instruct-2507",
           "name": "Qwen 3 235B A22B Instruct 2507",
@@ -37751,7 +37796,8 @@
           "name": "GLM-5.1",
           "cost": {
             "input": 1.4,
-            "output": 4.4
+            "output": 4.4,
+            "cache_read": 0.26
           },
           "context": 202752,
           "output_limit": 131072,
@@ -37822,7 +37868,8 @@
           "name": "Qwen3.7 Max",
           "cost": {
             "input": 1.25,
-            "output": 3.75
+            "output": 3.75,
+            "cache_read": 0.125
           },
           "context": 1000000,
           "output_limit": 500000,
@@ -37844,7 +37891,8 @@
           "name": "Qwen3.5 397B A17B",
           "cost": {
             "input": 0.6,
-            "output": 3.6
+            "output": 3.6,
+            "cache_read": 0.35
           },
           "context": 262144,
           "output_limit": 130000,
@@ -40988,6 +41036,21 @@
           "reasoning": true,
           "tool_call": true,
           "image": true
+        },
+        {
+          "id": "moonshotai/Kimi-K3",
+          "name": "Kimi K3",
+          "cost": {
+            "input": 3,
+            "output": 15,
+            "cache_read": 0.3
+          },
+          "context": 1048576,
+          "output_limit": 131072,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true,
+          "image": true
         }
       ]
     },
@@ -41448,7 +41511,9 @@
           "name": "MiniMax-M2.1",
           "cost": {
             "input": 2.1,
-            "output": 8.4
+            "output": 8.4,
+            "cache_read": 2.1,
+            "cache_write": 8.4
           },
           "context": 204800,
           "output_limit": 131072,
@@ -45670,7 +45735,9 @@
           "output_limit": 32000,
           "reasoning": true,
           "tool_call": true,
-          "structured_output": true
+          "structured_output": true,
+          "image": true,
+          "video": true
         },
         {
           "id": "nemotron-3-ultra",
@@ -48709,7 +48776,9 @@
           "name": "MiniMax-M2.1",
           "cost": {
             "input": 0.3,
-            "output": 1.2
+            "output": 1.2,
+            "cache_read": 0.03,
+            "cache_write": 0.375
           },
           "context": 204800,
           "output_limit": 131072,
@@ -49621,6 +49690,355 @@
         }
       ]
     },
+    "tensorx": {
+      "id": "tensorx",
+      "name": "TensorX",
+      "models": [
+        {
+          "id": "nvidia/nemotron-3-super-120b-a12b",
+          "name": "Nemotron 3 Super 120B A12B",
+          "cost": {
+            "input": 0.3,
+            "output": 0.9,
+            "cache_read": 0.075,
+            "cache_write": 0.375
+          },
+          "context": 262144,
+          "output_limit": 262144,
+          "reasoning": true,
+          "tool_call": true
+        },
+        {
+          "id": "qwen/qwen3-coder-30b-a3b-instruct",
+          "name": "Qwen3-Coder 30B-A3B Instruct",
+          "cost": {
+            "input": 0.06,
+            "output": 0.25,
+            "cache_read": 0.015,
+            "cache_write": 0.075
+          },
+          "context": 262000,
+          "output_limit": 65536,
+          "tool_call": true
+        },
+        {
+          "id": "qwen/qwen3.5-9b",
+          "name": "Qwen3.5 9B",
+          "cost": {
+            "input": 0.15,
+            "output": 0.2,
+            "cache_read": 0.0375,
+            "cache_write": 0.1875
+          },
+          "context": 262144,
+          "output_limit": 65536,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true,
+          "image": true,
+          "video": true
+        },
+        {
+          "id": "qwen/qwen3-vl-235b-a22b-instruct",
+          "name": "Qwen3 VL 235B-A22B Instruct",
+          "cost": {
+            "input": 0.21,
+            "output": 1.9,
+            "cache_read": 0.0525,
+            "cache_write": 0.2625
+          },
+          "context": 131000,
+          "output_limit": 131072,
+          "tool_call": true,
+          "image": true
+        },
+        {
+          "id": "qwen/qwen3-235b-a22b-2507",
+          "name": "Qwen3 235B-A22B-2507",
+          "cost": {
+            "input": 0.072,
+            "output": 0.464,
+            "cache_read": 0.018,
+            "cache_write": 0.09
+          },
+          "context": 131000,
+          "output_limit": 262144,
+          "tool_call": true
+        },
+        {
+          "id": "qwen/qwen3.5-122b-a10b",
+          "name": "Qwen3.5 122B-A10B",
+          "cost": {
+            "input": 0.5,
+            "output": 3.5,
+            "cache_read": 0.125,
+            "cache_write": 0.625
+          },
+          "context": 262144,
+          "output_limit": 262144,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true,
+          "image": true,
+          "video": true
+        },
+        {
+          "id": "minimax/minimax-m3",
+          "name": "MiniMax-M3",
+          "cost": {
+            "input": 0.4,
+            "output": 2,
+            "cache_read": 0.1
+          },
+          "context": 1048576,
+          "output_limit": 131072,
+          "reasoning": true,
+          "tool_call": true,
+          "image": true,
+          "video": true
+        },
+        {
+          "id": "minimax/minimax-m2.5",
+          "name": "MiniMax-M2.5",
+          "cost": {
+            "input": 0.3,
+            "output": 1.2,
+            "cache_read": 0.075,
+            "cache_write": 0.375
+          },
+          "context": 196608,
+          "output_limit": 65536,
+          "reasoning": true,
+          "tool_call": true
+        },
+        {
+          "id": "deepseek/deepseek-v4-flash",
+          "name": "DeepSeek V4 Flash",
+          "cost": {
+            "input": 0.15,
+            "output": 0.3,
+            "cache_read": 0.0375,
+            "cache_write": 0.1875
+          },
+          "context": 1048576,
+          "output_limit": 384000,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true
+        },
+        {
+          "id": "deepseek/deepseek-v4-pro",
+          "name": "DeepSeek V4 Pro",
+          "cost": {
+            "input": 1.75,
+            "output": 3.5,
+            "cache_read": 0.4375,
+            "cache_write": 2.185
+          },
+          "context": 1048576,
+          "output_limit": 384000,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true
+        },
+        {
+          "id": "deepseek/deepseek-r1-0528",
+          "name": "DeepSeek R1-0528",
+          "cost": {
+            "input": 0.66,
+            "output": 2.6,
+            "cache_read": 0.165,
+            "cache_write": 0.825
+          },
+          "context": 164000,
+          "output_limit": 8192,
+          "reasoning": true,
+          "tool_call": true
+        },
+        {
+          "id": "deepseek/deepseek-v3.2",
+          "name": "DeepSeek V3.2",
+          "cost": {
+            "input": 0.3,
+            "output": 0.5,
+            "cache_read": 0.075,
+            "cache_write": 0.375
+          },
+          "context": 163840,
+          "output_limit": 163840,
+          "reasoning": true,
+          "tool_call": true
+        },
+        {
+          "id": "deepseek/deepseek-chat-v3.1",
+          "name": "DeepSeek Chat V3.1",
+          "cost": {
+            "input": 0.2,
+            "output": 0.8,
+            "cache_read": 0.05,
+            "cache_write": 0.25
+          },
+          "context": 164000,
+          "output_limit": 163840,
+          "reasoning": true,
+          "tool_call": true
+        },
+        {
+          "id": "z-ai/glm-5",
+          "name": "GLM-5",
+          "cost": {
+            "input": 1,
+            "output": 3.2,
+            "cache_read": 0.25,
+            "cache_write": 1.25
+          },
+          "context": 202752,
+          "output_limit": 202752,
+          "reasoning": true,
+          "tool_call": true
+        },
+        {
+          "id": "z-ai/glm-5.1",
+          "name": "GLM-5.1",
+          "cost": {
+            "input": 1.4,
+            "output": 4.4,
+            "cache_read": 0.35,
+            "cache_write": 1.75
+          },
+          "context": 202752,
+          "output_limit": 202752,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true
+        },
+        {
+          "id": "z-ai/glm-5.2",
+          "name": "GLM-5.2",
+          "cost": {
+            "input": 1.5,
+            "output": 4.5,
+            "cache_read": 0.375
+          },
+          "context": 1048576,
+          "output_limit": 131072,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true
+        },
+        {
+          "id": "z-ai/glm-4.7",
+          "name": "GLM-4.7",
+          "cost": {
+            "input": 0.6,
+            "output": 2.2,
+            "cache_read": 0.15,
+            "cache_write": 0.75
+          },
+          "context": 200000,
+          "output_limit": 200000,
+          "reasoning": true,
+          "tool_call": true,
+          "image": true
+        },
+        {
+          "id": "z-ai/glm-5-turbo",
+          "name": "GLM-5-Turbo",
+          "cost": {
+            "input": 1.2,
+            "output": 4,
+            "cache_read": 0.3,
+            "cache_write": 1.5
+          },
+          "context": 202752,
+          "output_limit": 131072,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true
+        },
+        {
+          "id": "z-ai/glm-5v-turbo",
+          "name": "GLM-5V-Turbo",
+          "cost": {
+            "input": 1.2,
+            "output": 4,
+            "cache_read": 0.3,
+            "cache_write": 1.5
+          },
+          "context": 202752,
+          "output_limit": 131072,
+          "reasoning": true,
+          "tool_call": true,
+          "image": true,
+          "video": true
+        },
+        {
+          "id": "moonshotai/kimi-k2.5",
+          "name": "Kimi K2.5",
+          "cost": {
+            "input": 0.5,
+            "output": 2.8,
+            "cache_read": 0.125,
+            "cache_write": 0.625
+          },
+          "context": 262144,
+          "output_limit": 262144,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true,
+          "image": true,
+          "video": true
+        },
+        {
+          "id": "moonshotai/kimi-k2.6",
+          "name": "Kimi K2.6",
+          "cost": {
+            "input": 1,
+            "output": 4,
+            "cache_read": 0.25,
+            "cache_write": 1.25
+          },
+          "context": 262144,
+          "output_limit": 262144,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true,
+          "image": true,
+          "video": true
+        },
+        {
+          "id": "moonshotai/kimi-k2.7-code",
+          "name": "Kimi K2.7 Code",
+          "cost": {
+            "input": 1.25,
+            "output": 4.5,
+            "cache_read": 0.3125
+          },
+          "context": 262144,
+          "output_limit": 262144,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true,
+          "image": true,
+          "video": true
+        },
+        {
+          "id": "openai/gpt-oss-120b",
+          "name": "GPT OSS 120B",
+          "cost": {
+            "input": 0.04,
+            "output": 0.2,
+            "cache_read": 0.01,
+            "cache_write": 0.05
+          },
+          "context": 131072,
+          "output_limit": 32768,
+          "reasoning": true,
+          "tool_call": true,
+          "structured_output": true
+        }
+      ]
+    },
     "llama": {
       "id": "llama",
       "name": "Llama",
@@ -58117,7 +58535,8 @@
             "cache_read": 0.03
           },
           "context": 32768,
-          "output_limit": 65536
+          "output_limit": 65536,
+          "image": true
         },
         {
           "id": "google/imagen-4.0-generate-001",
@@ -58133,7 +58552,8 @@
             "cache_read": 0.2
           },
           "context": 65536,
-          "output_limit": 32768
+          "output_limit": 32768,
+          "image": true
         },
         {
           "id": "google/gemini-3.1-pro-preview",
@@ -58150,6 +58570,10 @@
           "structured_output": true,
           "image": true
         },
+        {
+          "id": "google/veo-3.1-lite-generate-001",
+          "name": "Veo 3.1 Lite Generate"
+        },
         {
           "id": "google/gemini-2.5-flash-lite",
           "name": "Gemini 2.5 Flash Lite",
@@ -58868,7 +59292,8 @@
           },
           "context": 128000,
           "output_limit": 128000,
-          "tool_call": true
+          "tool_call": true,
+          "image": true
         },
         {
           "id": "mistral/mistral-small",
@@ -59198,6 +59623,19 @@
           "tool_call": true,
           "structured_output": true
         },
+        {
+          "id": "deepseek/deepseek-v4-flash-0731",
+          "name": "DeepSeek V4 Flash 0731",
+          "cost": {
+            "input": 0.13,
+            "output": 0.26,
+            "cache_read": 0.028
+          },
+          "context": 1000000,
+          "output_limit": 384000,
+          "reasoning": true,
+          "tool_call": true
+        },
         {
           "id": "deepseek/deepseek-v3",
           "name": "DeepSeek V3 0324",
@@ -59714,7 +60152,8 @@
           "context": 256000,
           "output_limit": 80000,
           "reasoning": true,
-          "tool_call": true
+          "tool_call": true,
+          "image": true
         },
         {
           "id": "kwaipilot/kat-coder-air-v2.5",
@@ -59727,7 +60166,8 @@
           "context": 256000,
           "output_limit": 80000,
           "reasoning": true,
-          "tool_call": true
+          "tool_call": true,
+          "image": true
         },
         {
           "id": "xiaomi/mimo-v2.5",
@@ -60018,7 +60458,8 @@
           "name": "Sonar Reasoning Pro",
           "context": 127000,
           "output_limit": 8000,
-          "reasoning": true
+          "reasoning": true,
+          "image": true
         },
         {
           "id": "moonshotai/kimi-k2.5",
@@ -61102,7 +61543,8 @@
           "context": 262114,
           "output_limit": 262114,
           "reasoning": true,
-          "tool_call": true
+          "tool_call": true,
+          "image": true
         },
         {
           "id": "stepfun/step-3.7-flash",
@@ -68833,7 +69275,9 @@
           "name": "MiniMax-M2.1",
           "cost": {
             "input": 0,
-            "output": 0
+            "output": 0,
+            "cache_read": 0,
+            "cache_write": 0
           },
           "context": 204800,
           "output_limit": 131072,
@@ -68910,6 +69354,7 @@
           "cost": {
             "input": 1.4,
             "output": 4.4,
+            "cache_read": 0.26,
             "cache_write": 0
           },
           "context": 1049000,
@@ -68973,6 +69418,7 @@
           "cost": {
             "input": 1.4,
             "output": 4.4,
+            "cache_read": 0.26,
             "cache_write": 0
           },
           "context": 205000,
@@ -69064,7 +69510,8 @@
           "name": "Pro/moonshotai/Kimi-K2.5",
           "cost": {
             "input": 0.45,
-            "output": 2.25
+            "output": 2.25,
+            "cache_read": 0.07
           },
           "context": 262000,
           "output_limit": 262000,
@@ -72114,20 +72561,6 @@
           "structured_output": true,
           "image": true
         },
-        {
-          "id": "zai-org/GLM-5-TEE",
-          "name": "GLM 5 TEE",
-          "cost": {
-            "input": 0.95,
-            "output": 2.55,
-            "cache_read": 0.475
-          },
-          "context": 202752,
-          "output_limit": 65535,
-          "reasoning": true,
-          "tool_call": true,
-          "structured_output": true
-        },
         {
           "id": "zai-org/GLM-5.1-TEE",
           "name": "GLM 5.1 TEE",
@@ -72225,20 +72658,6 @@
           "structured_output": true,
           "image": true
         },
-        {
-          "id": "MiniMaxAI/MiniMax-M2.5-TEE",
-          "name": "MiniMax M2.5 TEE",
-          "cost": {
-            "input": 0.15,
-            "output": 1.2,
-            "cache_read": 0.075
-          },
-          "context": 196608,
-          "output_limit": 65536,
-          "reasoning": true,
-          "tool_call": true,
-          "structured_output": true
-        },
         {
           "id": "deepseek-ai/DeepSeek-V3.2-TEE",
           "name": "DeepSeek V3.2 TEE",
@@ -72269,22 +72688,6 @@
           "image": true,
           "video": true
         },
-        {
-          "id": "moonshotai/Kimi-K2.5-TEE",
-          "name": "Kimi K2.5 TEE",
-          "cost": {
-            "input": 0.44,
-            "output": 2,
-            "cache_read": 0.22
-          },
-          "context": 262144,
-          "output_limit": 65535,
-          "reasoning": true,
-          "tool_call": true,
-          "structured_output": true,
-          "image": true,
-          "video": true
-        },
         {
           "id": "moonshotai/Kimi-K3-TEE",
           "name": "Kimi K3 TEE",
@@ -77090,7 +77493,9 @@
           "name": "MiniMax-M2.1",
           "cost": {
             "input": 0.3,
-            "output": 1.2
+            "output": 1.2,
+            "cache_read": 0.03,
+            "cache_write": 0.375
           },
           "context": 204800,
           "output_limit": 131072,
@@ -77719,7 +78124,9 @@
           "name": "MiniMax-M2.1",
           "cost": {
             "input": 0,
-            "output": 0
+            "output": 0,
+            "cache_read": 0,
+            "cache_write": 0
           },
           "context": 204800,
           "output_limit": 131072,

src/lib/models-dev.json

@@ -13610,6 +13610,8 @@
     "reasoning": true,
     "tool_call": true,
     "structured_output": true,
+    "image": true,
+    "video": true,
     "cost": {
       "input": 0,
       "output": 0
@@ -13674,7 +13676,8 @@
     "structured_output": true,
     "cost": {
       "input": 5,
-      "output": 5
+      "output": 5,
+      "cache_read": 0.5
     }
   },
   "abliteration-ai/abliterated-model": {
@@ -13685,7 +13688,8 @@
     "image": true,
     "cost": {
       "input": 3,
-      "output": 3
+      "output": 3,
+      "cache_read": 0.3
     }
   },
   "alibaba-coding-plan-cn/qwen3.7-plus": {
@@ -14520,6 +14524,8 @@
     "reasoning": true,
     "tool_call": true,
     "structured_output": true,
+    "image": true,
+    "video": true,
     "cost": {
       "input": 0.1,
       "output": 0.15
@@ -15883,6 +15889,20 @@
       "cache_read": 0.2
     }
   },
+  "llmgateway/kimi-k3-fast": {
+    "name": "Kimi K3",
+    "context": 1040384,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "image": true,
+    "video": true,
+    "cost": {
+      "input": 4.5,
+      "output": 22.5,
+      "cache_read": 0.45
+    }
+  },
   "llmgateway/claude-opus-5": {
     "name": "Claude Opus 5",
     "context": 1000000,
@@ -23525,6 +23545,8 @@
     "reasoning": true,
     "tool_call": true,
     "structured_output": true,
+    "image": true,
+    "video": true,
     "cost": {
       "input": 0.3,
       "output": 0.3,
@@ -23853,7 +23875,8 @@
     "structured_output": true,
     "cost": {
       "input": 0.95,
-      "output": 2.55
+      "output": 2.55,
+      "cache_read": 0.2
     }
   },
   "siliconflow/zai-org/GLM-4.5-Air": {
@@ -23875,6 +23898,7 @@
     "cost": {
       "input": 1.4,
       "output": 4.4,
+      "cache_read": 0.26,
       "cache_write": 0
     }
   },
@@ -23887,6 +23911,7 @@
     "cost": {
       "input": 1.4,
       "output": 4.4,
+      "cache_read": 0.26,
       "cache_write": 0
     }
   },
@@ -23899,6 +23924,7 @@
     "cost": {
       "input": 1.2,
       "output": 4,
+      "cache_read": 0.24,
       "cache_write": 0
     }
   },
@@ -24174,7 +24200,8 @@
     "tool_call": true,
     "cost": {
       "input": 0.3,
-      "output": 1.2
+      "output": 1.2,
+      "cache_read": 0.03
     }
   },
   "siliconflow/inclusionAI/Ling-flash-2.0": {
@@ -24319,7 +24346,8 @@
     "image": true,
     "cost": {
       "input": 0.45,
-      "output": 2.25
+      "output": 2.25,
+      "cache_read": 0.07
     }
   },
   "siliconflow/openai/gpt-oss-20b": {
@@ -29988,8 +30016,9 @@
     "tool_call": true,
     "structured_output": true,
     "cost": {
-      "input": 0.125,
+      "input": 0.06,
-      "output": 0.5
+      "output": 0.4,
+      "cache_read": 0.01
     }
   },
   "venice/llama-3.2-3b": {
@@ -30013,6 +30042,18 @@
       "output": 225
     }
   },
+  "venice/deepseek-v4-flash-0731": {
+    "name": "DeepSeek V4 Flash 0731",
+    "context": 1000000,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "cost": {
+      "input": 0.175,
+      "output": 0.35,
+      "cache_read": 0.035
+    }
+  },
   "venice/qwen3-235b-a22b-instruct-2507": {
     "name": "Qwen 3 235B A22B Instruct 2507",
     "context": 128000,
@@ -30792,7 +30833,8 @@
     "structured_output": true,
     "cost": {
       "input": 1.4,
-      "output": 4.4
+      "output": 4.4,
+      "cache_read": 0.26
     }
   },
   "togetherai/pearl-ai/gemma-4-31b-it": {
@@ -30849,7 +30891,8 @@
     "tool_call": true,
     "cost": {
       "input": 1.25,
-      "output": 3.75
+      "output": 3.75,
+      "cache_read": 0.125
     }
   },
   "togetherai/Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
@@ -30869,7 +30912,8 @@
     "image": true,
     "cost": {
       "input": 0.6,
-      "output": 3.6
+      "output": 3.6,
+      "cache_read": 0.35
     }
   },
   "togetherai/Qwen/Qwen3.6-Plus": {
@@ -33439,6 +33483,19 @@
       "cache_read": 0.27
     }
   },
+  "modal/moonshotai/Kimi-K3": {
+    "name": "Kimi K3",
+    "context": 1048576,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "image": true,
+    "cost": {
+      "input": 3,
+      "output": 15,
+      "cache_read": 0.3
+    }
+  },
   "qihang-ai/gemini-2.5-flash": {
     "name": "Gemini 2.5 Flash",
     "context": 1048576,
@@ -33814,7 +33871,9 @@
     "tool_call": true,
     "cost": {
       "input": 2.1,
-      "output": 8.4
+      "output": 8.4,
+      "cache_read": 2.1,
+      "cache_write": 8.4
     }
   },
   "moark/GLM-4.7": {
@@ -37298,6 +37357,8 @@
     "reasoning": true,
     "tool_call": true,
     "structured_output": true,
+    "image": true,
+    "video": true,
     "cost": {
       "input": 0.16,
       "output": 0.48
@@ -39859,7 +39920,9 @@
     "tool_call": true,
     "cost": {
       "input": 0.3,
-      "output": 1.2
+      "output": 1.2,
+      "cache_read": 0.03,
+      "cache_write": 0.375
     }
   },
   "minimax/MiniMax-M2.5": {
@@ -40611,6 +40674,303 @@
       "cache_read": 0.11
     }
   },
+  "tensorx/nvidia/nemotron-3-super-120b-a12b": {
+    "name": "Nemotron 3 Super 120B A12B",
+    "context": 262144,
+    "reasoning": true,
+    "tool_call": true,
+    "cost": {
+      "input": 0.3,
+      "output": 0.9,
+      "cache_read": 0.075,
+      "cache_write": 0.375
+    }
+  },
+  "tensorx/qwen/qwen3-coder-30b-a3b-instruct": {
+    "name": "Qwen3-Coder 30B-A3B Instruct",
+    "context": 262000,
+    "tool_call": true,
+    "cost": {
+      "input": 0.06,
+      "output": 0.25,
+      "cache_read": 0.015,
+      "cache_write": 0.075
+    }
+  },
+  "tensorx/qwen/qwen3.5-9b": {
+    "name": "Qwen3.5 9B",
+    "context": 262144,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "image": true,
+    "video": true,
+    "cost": {
+      "input": 0.15,
+      "output": 0.2,
+      "cache_read": 0.0375,
+      "cache_write": 0.1875
+    }
+  },
+  "tensorx/qwen/qwen3-vl-235b-a22b-instruct": {
+    "name": "Qwen3 VL 235B-A22B Instruct",
+    "context": 131000,
+    "tool_call": true,
+    "image": true,
+    "cost": {
+      "input": 0.21,
+      "output": 1.9,
+      "cache_read": 0.0525,
+      "cache_write": 0.2625
+    }
+  },
+  "tensorx/qwen/qwen3-235b-a22b-2507": {
+    "name": "Qwen3 235B-A22B-2507",
+    "context": 131000,
+    "tool_call": true,
+    "cost": {
+      "input": 0.072,
+      "output": 0.464,
+      "cache_read": 0.018,
+      "cache_write": 0.09
+    }
+  },
+  "tensorx/qwen/qwen3.5-122b-a10b": {
+    "name": "Qwen3.5 122B-A10B",
+    "context": 262144,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "image": true,
+    "video": true,
+    "cost": {
+      "input": 0.5,
+      "output": 3.5,
+      "cache_read": 0.125,
+      "cache_write": 0.625
+    }
+  },
+  "tensorx/minimax/minimax-m3": {
+    "name": "MiniMax-M3",
+    "context": 1048576,
+    "reasoning": true,
+    "tool_call": true,
+    "image": true,
+    "video": true,
+    "cost": {
+      "input": 0.4,
+      "output": 2,
+      "cache_read": 0.1
+    }
+  },
+  "tensorx/minimax/minimax-m2.5": {
+    "name": "MiniMax-M2.5",
+    "context": 196608,
+    "reasoning": true,
+    "tool_call": true,
+    "cost": {
+      "input": 0.3,
+      "output": 1.2,
+      "cache_read": 0.075,
+      "cache_write": 0.375
+    }
+  },
+  "tensorx/deepseek/deepseek-v4-flash": {
+    "name": "DeepSeek V4 Flash",
+    "context": 1048576,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "cost": {
+      "input": 0.15,
+      "output": 0.3,
+      "cache_read": 0.0375,
+      "cache_write": 0.1875
+    }
+  },
+  "tensorx/deepseek/deepseek-v4-pro": {
+    "name": "DeepSeek V4 Pro",
+    "context": 1048576,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "cost": {
+      "input": 1.75,
+      "output": 3.5,
+      "cache_read": 0.4375,
+      "cache_write": 2.185
+    }
+  },
+  "tensorx/deepseek/deepseek-r1-0528": {
+    "name": "DeepSeek R1-0528",
+    "context": 164000,
+    "reasoning": true,
+    "tool_call": true,
+    "cost": {
+      "input": 0.66,
+      "output": 2.6,
+      "cache_read": 0.165,
+      "cache_write": 0.825
+    }
+  },
+  "tensorx/deepseek/deepseek-v3.2": {
+    "name": "DeepSeek V3.2",
+    "context": 163840,
+    "reasoning": true,
+    "tool_call": true,
+    "cost": {
+      "input": 0.3,
+      "output": 0.5,
+      "cache_read": 0.075,
+      "cache_write": 0.375
+    }
+  },
+  "tensorx/deepseek/deepseek-chat-v3.1": {
+    "name": "DeepSeek Chat V3.1",
+    "context": 164000,
+    "reasoning": true,
+    "tool_call": true,
+    "cost": {
+      "input": 0.2,
+      "output": 0.8,
+      "cache_read": 0.05,
+      "cache_write": 0.25
+    }
+  },
+  "tensorx/z-ai/glm-5": {
+    "name": "GLM-5",
+    "context": 202752,
+    "reasoning": true,
+    "tool_call": true,
+    "cost": {
+      "input": 1,
+      "output": 3.2,
+      "cache_read": 0.25,
+      "cache_write": 1.25
+    }
+  },
+  "tensorx/z-ai/glm-5.1": {
+    "name": "GLM-5.1",
+    "context": 202752,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "cost": {
+      "input": 1.4,
+      "output": 4.4,
+      "cache_read": 0.35,
+      "cache_write": 1.75
+    }
+  },
+  "tensorx/z-ai/glm-5.2": {
+    "name": "GLM-5.2",
+    "context": 1048576,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "cost": {
+      "input": 1.5,
+      "output": 4.5,
+      "cache_read": 0.375
+    }
+  },
+  "tensorx/z-ai/glm-4.7": {
+    "name": "GLM-4.7",
+    "context": 200000,
+    "reasoning": true,
+    "tool_call": true,
+    "image": true,
+    "cost": {
+      "input": 0.6,
+      "output": 2.2,
+      "cache_read": 0.15,
+      "cache_write": 0.75
+    }
+  },
+  "tensorx/z-ai/glm-5-turbo": {
+    "name": "GLM-5-Turbo",
+    "context": 202752,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "cost": {
+      "input": 1.2,
+      "output": 4,
+      "cache_read": 0.3,
+      "cache_write": 1.5
+    }
+  },
+  "tensorx/z-ai/glm-5v-turbo": {
+    "name": "GLM-5V-Turbo",
+    "context": 202752,
+    "reasoning": true,
+    "tool_call": true,
+    "image": true,
+    "video": true,
+    "cost": {
+      "input": 1.2,
+      "output": 4,
+      "cache_read": 0.3,
+      "cache_write": 1.5
+    }
+  },
+  "tensorx/moonshotai/kimi-k2.5": {
+    "name": "Kimi K2.5",
+    "context": 262144,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "image": true,
+    "video": true,
+    "cost": {
+      "input": 0.5,
+      "output": 2.8,
+      "cache_read": 0.125,
+      "cache_write": 0.625
+    }
+  },
+  "tensorx/moonshotai/kimi-k2.6": {
+    "name": "Kimi K2.6",
+    "context": 262144,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "image": true,
+    "video": true,
+    "cost": {
+      "input": 1,
+      "output": 4,
+      "cache_read": 0.25,
+      "cache_write": 1.25
+    }
+  },
+  "tensorx/moonshotai/kimi-k2.7-code": {
+    "name": "Kimi K2.7 Code",
+    "context": 262144,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "image": true,
+    "video": true,
+    "cost": {
+      "input": 1.25,
+      "output": 4.5,
+      "cache_read": 0.3125
+    }
+  },
+  "tensorx/openai/gpt-oss-120b": {
+    "name": "GPT OSS 120B",
+    "context": 131072,
+    "reasoning": true,
+    "tool_call": true,
+    "structured_output": true,
+    "cost": {
+      "input": 0.04,
+      "output": 0.2,
+      "cache_read": 0.01,
+      "cache_write": 0.05
+    }
+  },
   "llama/cerebras-llama-4-maverick-17b-128e-instruct": {
     "name": "Cerebras-Llama-4-Maverick-17B-128E-Instruct",
     "context": 128000,
@@ -47689,6 +48049,7 @@
   "vercel/google/gemini-2.5-flash-image": {
     "name": "Nano Banana (Gemini 2.5 Flash Image)",
     "context": 32768,
+    "image": true,
     "cost": {
       "input": 0.3,
       "output": 2.5,
@@ -47702,6 +48063,7 @@
   "vercel/google/gemini-3-pro-image": {
     "name": "Nano Banana Pro",
     "context": 65536,
+    "image": true,
     "cost": {
       "input": 2,
       "output": 12,
@@ -47721,6 +48083,9 @@
       "cache_read": 0.2
     }
   },
+  "vercel/google/veo-3.1-lite-generate-001": {
+    "name": "Veo 3.1 Lite Generate"
+  },
   "vercel/google/gemini-2.5-flash-lite": {
     "name": "Gemini 2.5 Flash Lite",
     "context": 1048576,
@@ -48298,6 +48663,7 @@
     "name": "Mistral Nemo",
     "context": 128000,
     "tool_call": true,
+    "image": true,
     "cost": {
       "input": 0.15,
       "output": 0.15
@@ -48578,6 +48944,17 @@
       "cache_read": 0.0036
     }
   },
+  "vercel/deepseek/deepseek-v4-flash-0731": {
+    "name": "DeepSeek V4 Flash 0731",
+    "context": 1000000,
+    "reasoning": true,
+    "tool_call": true,
+    "cost": {
+      "input": 0.13,
+      "output": 0.26,
+      "cache_read": 0.028
+    }
+  },
   "vercel/deepseek/deepseek-v3": {
     "name": "DeepSeek V3 0324",
     "context": 163840,
@@ -49002,6 +49379,7 @@
     "context": 256000,
     "reasoning": true,
     "tool_call": true,
+    "image": true,
     "cost": {
       "input": 0.74,
       "output": 2.96,
@@ -49013,6 +49391,7 @@
     "context": 256000,
     "reasoning": true,
     "tool_call": true,
+    "image": true,
     "cost": {
       "input": 0.15,
       "output": 0.6,
@@ -49266,7 +49645,8 @@
   "vercel/perplexity/sonar-reasoning-pro": {
     "name": "Sonar Reasoning Pro",
     "context": 127000,
-    "reasoning": true
+    "reasoning": true,
+    "image": true
   },
   "vercel/moonshotai/kimi-k2.5": {
     "name": "Kimi K2.5",
@@ -50157,6 +50537,7 @@
     "context": 262114,
     "reasoning": true,
     "tool_call": true,
+    "image": true,
     "cost": {
       "input": 0.09,
       "output": 0.3,
@@ -56667,7 +57048,9 @@
     "tool_call": true,
     "cost": {
       "input": 0,
-      "output": 0
+      "output": 0,
+      "cache_read": 0,
+      "cache_write": 0
     }
   },
   "minimax-coding-plan/MiniMax-M2.5": {
@@ -56729,6 +57112,7 @@
     "cost": {
       "input": 1.4,
       "output": 4.4,
+      "cache_read": 0.26,
       "cache_write": 0
     }
   },
@@ -56782,6 +57166,7 @@
     "cost": {
       "input": 1.4,
       "output": 4.4,
+      "cache_read": 0.26,
       "cache_write": 0
     }
   },
@@ -56860,7 +57245,8 @@
     "image": true,
     "cost": {
       "input": 0.45,
-      "output": 2.25
+      "output": 2.25,
+      "cache_read": 0.07
     }
   },
   "siliconflow-cn/Qwen/Qwen3.5-27B": {
@@ -59386,18 +59772,6 @@
       "cache_read": 0.06
     }
   },
-  "chutes/zai-org/GLM-5-TEE": {
-    "name": "GLM 5 TEE",
-    "context": 202752,
-    "reasoning": true,
-    "tool_call": true,
-    "structured_output": true,
-    "cost": {
-      "input": 0.95,
-      "output": 2.55,
-      "cache_read": 0.475
-    }
-  },
   "chutes/zai-org/GLM-5.1-TEE": {
     "name": "GLM 5.1 TEE",
     "context": 202752,
@@ -59481,18 +59855,6 @@
       "cache_read": 0.225
     }
   },
-  "chutes/MiniMaxAI/MiniMax-M2.5-TEE": {
-    "name": "MiniMax M2.5 TEE",
-    "context": 196608,
-    "reasoning": true,
-    "tool_call": true,
-    "structured_output": true,
-    "cost": {
-      "input": 0.15,
-      "output": 1.2,
-      "cache_read": 0.075
-    }
-  },
   "chutes/deepseek-ai/DeepSeek-V3.2-TEE": {
     "name": "DeepSeek V3.2 TEE",
     "context": 131072,
@@ -59519,20 +59881,6 @@
       "cache_read": 0.33
     }
   },
-  "chutes/moonshotai/Kimi-K2.5-TEE": {
-    "name": "Kimi K2.5 TEE",
-    "context": 262144,
-    "reasoning": true,
-    "tool_call": true,
-    "structured_output": true,
-    "image": true,
-    "video": true,
-    "cost": {
-      "input": 0.44,
-      "output": 2,
-      "cache_read": 0.22
-    }
-  },
   "chutes/moonshotai/Kimi-K3-TEE": {
     "name": "Kimi K3 TEE",
     "context": 1048576,
@@ -63413,7 +63761,9 @@
     "tool_call": true,
     "cost": {
       "input": 0.3,
-      "output": 1.2
+      "output": 1.2,
+      "cache_read": 0.03,
+      "cache_write": 0.375
     }
   },
   "minimax-cn/MiniMax-M2.5": {
@@ -63930,7 +64280,9 @@
     "tool_call": true,
     "cost": {
       "input": 0,
-      "output": 0
+      "output": 0,
+      "cache_read": 0,
+      "cache_write": 0
     }
   },
   "minimax-cn-coding-plan/MiniMax-M2.5": {

website/index.html

@@ -13,7 +13,7 @@
     />
     <meta
       name="keywords"
-      content="KimiSwitch, KimiCodeSwitch, SwitchKimi, Kimi Code Switch, Kimi 切换, LLM 供应商管理, AI 供应商切换, 多 LLM 管理, Kimi 用量监控, 智谱 GLM, MiniMax, DeepSeek, OpenRouter, 基元律动, 用量查询, 账单监控, Token 套餐, Claude, GPT, Windows LLM 桌面工具"
+      content="Kimi Switch, KimiSwitch, KimiCode Switch, KimiCodeSwitch, SwitchKimi, Kimi Code Switch, Kimi 切换, LLM 供应商管理, AI 供应商切换, 多 LLM 管理, Kimi 用量监控, 智谱 GLM, MiniMax, DeepSeek, OpenRouter, 基元律动, 用量查询, 账单监控, Token 套餐, Claude, GPT, Windows LLM 桌面工具"
     />
 
     <!-- Open Graph -->
@@ -46,7 +46,7 @@
           {
             "@type": "SoftwareApplication",
             "name": "Kimi Switch",
-            "alternateName": ["KimiSwitch", "KimiCodeSwitch"],
+            "alternateName": ["KimiSwitch", "KimiCodeSwitch", "Kimi Switch", "KimiCode Switch"],
             "url": "https://billowliu2.github.io/KimiSwitch/",
             "applicationCategory": "DeveloperApplication",
             "operatingSystem": "Windows 10, Windows 11",

website/public/logos/glm.svg

@@ -1 +0,0 @@
-<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="#0F62FE"><path d="M12 2L4 6v6c0 5 3.5 8.5 8 10 4.5-1.5 8-5 8-10V6l-8-4zm0 4l4 2v4c0 3-1.5 5-4 6-2.5-1-4-3-4-6V8l4-2z"/></svg>

website/src/components/Footer.tsx

@@ -5,7 +5,7 @@ import { asset } from "../lib/asset";
 /** Map recommendLink.name → logo file in public/logos (only when we have one). */
 const logoFor: Record<string, string> = {
   "Kimi": "logos/kimi.svg",
-  "智谱 GLM": "logos/glm.svg",
+  "智谱 GLM": "logos/zhipu.svg",
   "DeepSeek": "logos/deepseek.svg",
   "OpenCodeGo": "logos/opencode.svg",
   "MiniMax": "logos/minimax.svg",

website/src/i18n.tsx

@@ -16,10 +16,10 @@ const zh = {
     performance: "性能",
     changelog: "更新日志",
     download: "下载",
-    downloadBtn: "下载 v0.6.8",
+    downloadBtn: "下载 v0.6.9",
   },
   hero: {
-    badge: "开源 MIT · Windows 10 / 11",
+    badge: "v0.6.9 · 开源 MIT · Windows / macOS / Linux",
     titleBefore: "统一管理你的",
     titleAccent: "AI 供应商",
     titleAfter: "",
@@ -140,6 +140,15 @@ const zh = {
     syncedNote: "数据已同步 GitHub Releases",
     fallbackNote: "内置版本记录",
     entries: [
+      {
+        version: "v0.6.9",
+        date: "2026-08-01",
+        items: [
+          "最近请求新增缓存命中率列，每次请求命中一目了然",
+          "中文模式费用按最新汇率换算人民币（¥）显示，英文模式保持美元",
+          "仪表盘性能再优化：价格解析记忆化缓存，扫描从秒级降到百毫秒级；右上角新增加载耗时/数据量跟踪",
+        ],
+      },
       {
         version: "v0.6.8",
         date: "2026-08-01",
@@ -150,14 +159,6 @@ const zh = {
           "官网更新：Kimi 风格重设计、中英文切换、更新日志同步 GitHub Releases、三平台下载",
         ],
       },
-      {
-        version: "v0.6.7",
-        date: "2026-08-01",
-        items: [
-          "应用内置 Kimi 设备授权登录（等同 kimi login），浏览器授权即用，过期自动续期",
-          "外链打开统一走 Rust 命令，修复部分链接无法跳转",
-        ],
-      },
       {
         version: "v0.6.6",
         date: "2026-08-01",
@@ -244,7 +245,7 @@ const zh = {
   },
   download: {
     title: "下载 Kimi Switch",
-    subtitle: "当前版本 v0.6.8 · Windows / macOS / Linux 三平台已发布",
+    subtitle: "当前版本 v0.6.9 · Windows / macOS / Linux 三平台已发布",
     autoUpdate: "应用内置自动检测更新，新版本发布后在设置页一键升级。",
     ready: "已发布",
     wip: "开发中",
@@ -312,10 +313,10 @@ const en: Dict = {
     performance: "Performance",
     changelog: "Changelog",
     download: "Download",
-    downloadBtn: "Download v0.6.8",
+    downloadBtn: "Download v0.6.9",
   },
   hero: {
-    badge: "Open source MIT · Windows 10 / 11",
+    badge: "v0.6.9 · Open source MIT · Windows / macOS / Linux",
     titleBefore: "One app for all your ",
     titleAccent: "AI providers",
     titleAfter: "",
@@ -436,6 +437,15 @@ const en: Dict = {
     syncedNote: "Synced from GitHub Releases",
     fallbackNote: "Built-in release notes",
     entries: [
+      {
+        version: "v0.6.9",
+        date: "2026-08-01",
+        items: [
+          "Recent requests gain a cache-hit-rate column, per request at a glance",
+          "Chinese UI now shows costs in CNY (latest exchange rate); English keeps USD",
+          "Dashboard speedup: memoized price resolution drops the scan from seconds to milliseconds; load time / payload tracking in the top-right corner",
+        ],
+      },
       {
         version: "v0.6.8",
         date: "2026-08-01",
@@ -540,7 +550,7 @@ const en: Dict = {
   },
   download: {
     title: "Download Kimi Switch",
-    subtitle: "Current version v0.6.8 · Windows, macOS and Linux now released",
+    subtitle: "Current version v0.6.9 · Windows, macOS and Linux now released",
     autoUpdate: "Built-in update detection: upgrade in one click from Settings when a new version ships.",
     ready: "Available",
     wip: "In development",

website/src/sections/Changelog.tsx

@@ -10,9 +10,14 @@ interface ChangelogEntry {
 
 const API_URL =
   "https://api.github.com/repos/billowliu2/KimiSwitch/releases?per_page=20";
-const CACHE_KEY = "kimi-switch-changelog";
 const CACHE_TTL = 60 * 60 * 1000; // 1 hour
 
+/** Cache key is versioned by the newest embedded release so redeploys
+ *  invalidate stale caches in the browser immediately. */
+function cacheKey(embedded: ChangelogEntry[]) {
+  return `kimi-switch-changelog-${embedded[0]?.version ?? "v1"}`;
+}
+
 /** Parse a GitHub release body into bullet items. */
 function parseBody(body: string): string[] {
   return body
@@ -36,9 +41,9 @@ async function fetchReleases(): Promise<ChangelogEntry[]> {
   }));
 }
 
-function readCache(): ChangelogEntry[] | null {
+function readCache(embedded: ChangelogEntry[]): ChangelogEntry[] | null {
   try {
-    const raw = localStorage.getItem(CACHE_KEY);
+    const raw = localStorage.getItem(cacheKey(embedded));
     if (!raw) return null;
     const { ts, data } = JSON.parse(raw) as { ts: number; data: ChangelogEntry[] };
     if (Date.now() - ts > CACHE_TTL) return null;
@@ -48,10 +53,10 @@ function readCache(): ChangelogEntry[] | null {
   }
 }
 
-function writeCache(data: ChangelogEntry[]) {
+function writeCache(data: ChangelogEntry[], embedded: ChangelogEntry[]) {
   try {
     localStorage.setItem(
-      CACHE_KEY,
+      cacheKey(embedded),
       JSON.stringify({ ts: Date.now(), data }),
     );
   } catch {
@@ -67,7 +72,7 @@ export default function Changelog() {
 
   useEffect(() => {
     let cancelled = false;
-    const cached = readCache();
+    const cached = readCache(embedded);
     if (cached) {
       setRemote(cached);
       setSynced(true);
@@ -78,7 +83,7 @@ export default function Changelog() {
         if (cancelled) return;
         setRemote(data);
         setSynced(true);
-        writeCache(data);
+        writeCache(data, embedded);
       })
       .catch(() => {
         /* 网络不通时使用内置列表兜底 */

website/src/sections/Download.tsx

@@ -8,25 +8,25 @@ const icons: Record<string, Icon> = {
   linux: LinuxLogo,
 };
 
-const VERSION = "0.6.8";
+const VERSION = "0.6.9";
 const GITHUB = "https://github.com/billowliu2/KimiSwitch";
 const MIRROR_RELEASES = "https://git.codingplan.site/admin/KimiCodeSwitch/releases";
 const dl = (file: string) => `${GITHUB}/releases/download/v${VERSION}/${file}`;
 
 /** Per-platform download assets for the current release (names match the CI
- *  release workflow, e.g. Kimi.Switch_0.6.8_x64_en-US.msi). */
+ *  release workflow, e.g. Kimi.Switch_0.6.9_x64_en-US.msi). */
 const assets: Record<string, { label: string; href: string }[]> = {
   windows: [
-    { label: "MSI", href: dl("Kimi.Switch_0.6.8_x64_en-US.msi") },
+    { label: "MSI", href: dl("Kimi.Switch_0.6.9_x64_en-US.msi") },
     { label: "镜像", href: MIRROR_RELEASES },
   ],
   macos: [
-    { label: "Apple Silicon (.dmg)", href: dl("Kimi.Switch_0.6.8_aarch64.dmg") },
+    { label: "Apple Silicon (.dmg)", href: dl("Kimi.Switch_0.6.9_aarch64.dmg") },
   ],
   linux: [
-    { label: ".deb", href: dl("Kimi.Switch_0.6.8_amd64.deb") },
+    { label: ".deb", href: dl("Kimi.Switch_0.6.9_amd64.deb") },
-    { label: ".AppImage", href: dl("Kimi.Switch_0.6.8_amd64.AppImage") },
+    { label: ".AppImage", href: dl("Kimi.Switch_0.6.9_amd64.AppImage") },
-    { label: ".rpm", href: dl("Kimi.Switch-0.6.8-1.x86_64.rpm") },
+    { label: ".rpm", href: dl("Kimi.Switch-0.6.9-1.x86_64.rpm") },
   ],
 };