diff --git a/README.md b/README.md
index 75c0acd..6c27fb6 100644
--- a/README.md
+++ b/README.md
@@ -129,6 +129,8 @@ Playwright 不下载独立 Chromium。首次 `setup` 只缓存 MCP 的 Node.js 
 
 MCP 服务器可通过 Web UI 左下角的 MCP 按钮可视化配置（写入 `data/agent/mcp.json`）：支持 stdio（命令 + 参数）与 HTTP（URL + 请求头 + OAuth/Bearer）两种传输、环境变量键值编辑、工作目录、生命周期与超时设置，另保留原始 JSON 编辑兜底。保存后重启 pi（或 /reload）生效。
 
+Web UI 右上角的会话信息栏会汇总 Token 使用情况；当模型返回缓存读写数据时，还会显示按 Token 加权计算的缓存命中率：`cacheRead / (input + cacheRead + cacheWrite)`，不计输出 Token。
+
 #### 记忆模式
 
 默认设置 `PI_MEMORY_NO_SEARCH=1` 和 `PI_MEMORY_QMD_UPDATE=off`。这不会削弱 Markdown 记忆、每日日志、scratchpad 或恢复记录，但会跳过 qmd 探测、安装提示和 `memory_search` 工具，因此全新安装不会自动下载 qmd 的本地模型。集成补丁会在 `setup` 以及每次启动前自动检查并重放，受管插件重装后无需手工修改。
@@ -324,6 +326,8 @@ Playwright never downloads a standalone Chromium in this project. Setup caches o
 
 MCP servers can be configured visually from the MCP button in the lower-left Web UI (writes `data/agent/mcp.json`): stdio (command + args) or HTTP (URL + headers + OAuth/Bearer) transport, environment-variable row editing, working directory, lifecycle and timeout options, plus raw JSON editing as a fallback. Changes take effect after restarting pi (or /reload).
 
+The Web UI session-info panel in the upper-right summarizes Token usage. When a model reports cache read/write data, it also shows the token-weighted cache hit rate: `cacheRead / (input + cacheRead + cacheWrite)`, excluding output Tokens.
+
 #### Memory modes
 
 The managed launcher defaults to `PI_MEMORY_NO_SEARCH=1` and `PI_MEMORY_QMD_UPDATE=off`. Markdown memory, daily logs, scratchpad, and recovery stay available, while qmd detection, its installation notice, and the `memory_search` tool are skipped. A deterministic integration patch is checked during setup and before every launch, so reinstalling managed packages needs no manual repair.
diff --git a/pi-web/components/AppShell.tsx b/pi-web/components/AppShell.tsx
index 3aa6ef4..d45cdde 100644
--- a/pi-web/components/AppShell.tsx
+++ b/pi-web/components/AppShell.tsx
@@ -1256,11 +1256,16 @@ export function AppShell() {
                        [translate("session.toolResults"), sessionStats.toolResults.toLocaleString(locale)],
                        [translate("session.total"), sessionStats.totalMessages.toLocaleString(locale)],
                     ];
+                    const promptTokens = sessionStats.tokens.input + sessionStats.tokens.cacheRead + sessionStats.tokens.cacheWrite;
+                    const cacheHitRate = promptTokens > 0 && (sessionStats.tokens.cacheRead > 0 || sessionStats.tokens.cacheWrite > 0)
+                      ? `${((sessionStats.tokens.cacheRead / promptTokens) * 100).toFixed(1)}%`
+                      : null;
                     const tokenRows = [
                        [translate("session.input"), sessionStats.tokens.input.toLocaleString(locale)],
                        [translate("session.output"), sessionStats.tokens.output.toLocaleString(locale)],
                        ...(sessionStats.tokens.cacheRead > 0 ? [[translate("session.cacheRead"), sessionStats.tokens.cacheRead.toLocaleString(locale)]] : []),
                        ...(sessionStats.tokens.cacheWrite > 0 ? [[translate("session.cacheWrite"), sessionStats.tokens.cacheWrite.toLocaleString(locale)]] : []),
+                       ...(cacheHitRate ? [[translate("session.cacheHitRate"), cacheHitRate]] : []),
                        [translate("session.total"), sessionStats.tokens.total.toLocaleString(locale)],
                     ];
                     const ctx = contextUsage ?? sessionStats.contextUsage;
diff --git a/pi-web/lib/i18n/messages/en.ts b/pi-web/lib/i18n/messages/en.ts
index 556520a..0df965b 100644
--- a/pi-web/lib/i18n/messages/en.ts
+++ b/pi-web/lib/i18n/messages/en.ts
@@ -43,6 +43,7 @@ export const enLocale: LocalePlugin = {
     "session.output": "Output",
     "session.cacheRead": "Cache Read",
     "session.cacheWrite": "Cache Write",
+    "session.cacheHitRate": "Cache Hit Rate",
     "session.cost": "Cost",
     "session.context": "Context",
     "session.copy": "Copy {value}",
diff --git a/pi-web/lib/i18n/messages/zh-CN.ts b/pi-web/lib/i18n/messages/zh-CN.ts
index d6e2026..61d1fe8 100644
--- a/pi-web/lib/i18n/messages/zh-CN.ts
+++ b/pi-web/lib/i18n/messages/zh-CN.ts
@@ -43,6 +43,7 @@ export const zhCNLocale: LocalePlugin = {
     "session.output": "输出",
     "session.cacheRead": "缓存读取",
     "session.cacheWrite": "缓存写入",
+    "session.cacheHitRate": "缓存命中率",
     "session.cost": "费用",
     "session.context": "上下文",
     "session.copy": "复制{value}",