i/Aether
1
0
Fork 0
mirror of https://github.com/fawney19/Aether.git synced 2026-09-02 01:10:23 +08:00
Code Issues Packages Projects Releases Wiki Activity

feat(usage): 统计表格新增输入/输出与缓存创建token细分展示

Browse Source 
- 后端接口(admin/user_me/query)新增 output_tokens、cache_creation_tokens、total_input_context 字段返回
- 前端统计表格(模型/提供商/API格式)合并 Tokens 列为"输入/输出"+"缓存读取/创建"两行紧凑布局
- 合并"缓存Token"和"缓存命中率"为单一"命中率"列，减少表格宽度
- 修正 UsageRecordsTable 缩进及模板格式
This commit is contained in:
fawney19
2026-03-14 16:11:56 +08:00
parent aafd332198
commit 751a4d9111
11 changed files with 268 additions and 186 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
frontend/src/features/usage/components/UsageApiFormatTable.vue
View File

@@ -16,16 +16,16 @@
请求数 请求数
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
Tokens <div class="flex flex-col text-xs gap-0.5">
<span>输入/输出</span>
<span class="text-muted-foreground font-normal">缓存</span>
</div>
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
费用 费用
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
缓存Token 命中率
</TableHead>
<TableHead class="h-8 px-2 text-right">
缓存命中率
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
平均响应 平均响应
@@ -35,7 +35,7 @@
<TableBody> <TableBody>
<TableRow v-if="data.length === 0"> <TableRow v-if="data.length === 0">
<TableCell <TableCell
:colspan="7" :colspan="6"
class="text-center py-6 text-muted-foreground px-2" class="text-center py-6 text-muted-foreground px-2"
> >
暂无API格式统计数据 暂无API格式统计数据
@@ -52,7 +52,10 @@
{{ item.request_count }} {{ item.request_count }}
</TableCell> </TableCell>
<TableCell class="text-right py-2 px-2"> <TableCell class="text-right py-2 px-2">
<span>{{ formatTokens(item.total_tokens) }}</span> <div class="flex flex-col items-end text-xs gap-0.5">
<span>{{ formatTokens(item.total_input_context || 0) }} / {{ formatTokens(item.output_tokens || 0) }}</span>
<span class="text-muted-foreground">{{ formatTokens(item.cache_read_tokens || 0) }} / {{ formatTokens(item.cache_creation_tokens || 0) }}</span>
</div>
</TableCell> </TableCell>
<TableCell class="text-right py-2 px-2"> <TableCell class="text-right py-2 px-2">
<div class="flex flex-col items-end text-xs gap-0.5"> <div class="flex flex-col items-end text-xs gap-0.5">
@@ -66,10 +69,7 @@
</div> </div>
</TableCell> </TableCell>
<TableCell class="text-right py-2 px-2"> <TableCell class="text-right py-2 px-2">
{{ formatTokens(item.cache_read_tokens || 0) }} <span>{{ formatHitRate(item.cache_hit_rate) }}</span>
</TableCell>
<TableCell class="text-right py-2 px-2 text-muted-foreground">
{{ formatHitRate(item.cache_hit_rate) }}
</TableCell> </TableCell>
<TableCell class="text-right text-muted-foreground py-2 px-2"> <TableCell class="text-right text-muted-foreground py-2 px-2">
{{ item.avgResponseTime }} {{ item.avgResponseTime }}

22
frontend/src/features/usage/components/UsageModelTable.vue
View File

@@ -16,16 +16,16 @@
请求数 请求数
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
Tokens <div class="flex flex-col text-xs gap-0.5">
<span>输入/输出</span>
<span class="text-muted-foreground font-normal">缓存</span>
</div>
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
费用 费用
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
缓存Token 命中率
</TableHead>
<TableHead class="h-8 px-2 text-right">
缓存命中率
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
效率 效率
@@ -35,7 +35,7 @@
<TableBody> <TableBody>
<TableRow v-if="data.length === 0"> <TableRow v-if="data.length === 0">
<TableCell <TableCell
:colspan="7" :colspan="6"
class="text-center py-6 text-muted-foreground px-2" class="text-center py-6 text-muted-foreground px-2"
> >
暂无模型统计数据 暂无模型统计数据
@@ -52,7 +52,10 @@
{{ model.request_count }} {{ model.request_count }}
</TableCell> </TableCell>
<TableCell class="text-right py-2 px-2"> <TableCell class="text-right py-2 px-2">
<span>{{ formatTokens(model.total_tokens) }}</span> <div class="flex flex-col items-end text-xs gap-0.5">
<span>{{ formatTokens(model.total_input_context || 0) }} / {{ formatTokens(model.output_tokens || 0) }}</span>
<span class="text-muted-foreground">{{ formatTokens(model.cache_read_tokens || 0) }} / {{ formatTokens(model.cache_creation_tokens || 0) }}</span>
</div>
</TableCell> </TableCell>
<TableCell class="text-right py-2 px-2"> <TableCell class="text-right py-2 px-2">
<div class="flex flex-col items-end text-xs gap-0.5"> <div class="flex flex-col items-end text-xs gap-0.5">
@@ -66,10 +69,7 @@
</div> </div>
</TableCell> </TableCell>
<TableCell class="text-right py-2 px-2"> <TableCell class="text-right py-2 px-2">
{{ formatTokens(model.cache_read_tokens || 0) }} <span>{{ formatHitRate(model.cache_hit_rate) }}</span>
</TableCell>
<TableCell class="text-right py-2 px-2 text-muted-foreground">
{{ formatHitRate(model.cache_hit_rate) }}
</TableCell> </TableCell>
<TableCell class="text-right text-muted-foreground py-2 px-2"> <TableCell class="text-right text-muted-foreground py-2 px-2">
{{ model.costPerToken }} {{ model.costPerToken }}

23
frontend/src/features/usage/components/UsageProviderTable.vue
View File

@@ -16,16 +16,16 @@
请求数 请求数
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
Tokens <div class="flex flex-col text-xs gap-0.5">
<span>输入/输出</span>
<span class="text-muted-foreground font-normal">缓存</span>
</div>
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
费用 费用
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
缓存Token 命中率
</TableHead>
<TableHead class="h-8 px-2 text-right">
缓存命中率
</TableHead> </TableHead>
<TableHead class="h-8 px-2 text-right"> <TableHead class="h-8 px-2 text-right">
成功率 成功率
@@ -38,7 +38,7 @@
<TableBody> <TableBody>
<TableRow v-if="data.length === 0"> <TableRow v-if="data.length === 0">
<TableCell <TableCell
:colspan="8" :colspan="7"
class="text-center py-6 text-muted-foreground px-2" class="text-center py-6 text-muted-foreground px-2"
> >
暂无提供商统计数据 暂无提供商统计数据
@@ -55,7 +55,10 @@
{{ provider.requests }} {{ provider.requests }}
</TableCell> </TableCell>
<TableCell class="text-right py-2 px-2"> <TableCell class="text-right py-2 px-2">
<span>{{ formatTokens(provider.totalTokens) }}</span> <div class="flex flex-col items-end text-xs gap-0.5">
<span>{{ formatTokens(provider.totalInputContext || 0) }} / {{ formatTokens(provider.outputTokens || 0) }}</span>
<span class="text-muted-foreground">{{ formatTokens(provider.cacheReadTokens || 0) }} / {{ formatTokens(provider.cacheCreationTokens || 0) }}</span>
</div>
</TableCell> </TableCell>
<TableCell class="text-right py-2 px-2"> <TableCell class="text-right py-2 px-2">
<div class="flex flex-col items-end text-xs gap-0.5"> <div class="flex flex-col items-end text-xs gap-0.5">
@@ -69,10 +72,7 @@
</div> </div>
</TableCell> </TableCell>
<TableCell class="text-right py-2 px-2"> <TableCell class="text-right py-2 px-2">
{{ formatTokens(provider.cacheReadTokens || 0) }} <span>{{ formatHitRate(provider.cacheHitRate) }}</span>
</TableCell>
<TableCell class="text-right py-2 px-2 text-muted-foreground">
{{ formatHitRate(provider.cacheHitRate) }}
</TableCell> </TableCell>
<TableCell class="text-right py-2 px-2"> <TableCell class="text-right py-2 px-2">
<span :class="getSuccessRateClass(provider.successRate)">{{ provider.successRate }}%</span> <span :class="getSuccessRateClass(provider.successRate)">{{ provider.successRate }}%</span>
@@ -108,5 +108,4 @@ function getSuccessRateClass(rate: number): string {
if (rate < 90) return 'text-destructive' if (rate < 90) return 'text-destructive'
return '' // 默认颜色 return '' // 默认颜色
} }
</script> </script>

296
frontend/src/features/usage/components/UsageRecordsTable.vue
View File

@@ -3,167 +3,167 @@
<template #actions> <template #actions>
<!-- 时间范围筛选 --> <!-- 时间范围筛选 -->
<TimeRangePicker <TimeRangePicker
v-model="timeRangeModel" v-model="timeRangeModel"
:show-granularity="false" :show-granularity="false"
/> />
<!-- 分隔线 --> <!-- 分隔线 -->
<div class="hidden sm:block h-4 w-px bg-border" /> <div class="hidden sm:block h-4 w-px bg-border" />
<!-- 通用搜索 --> <!-- 通用搜索 -->
<div class="relative"> <div class="relative">
<Search class="absolute left-2.5 top-1/2 -translate-y-1/2 h-3.5 w-3.5 text-muted-foreground z-10 pointer-events-none" /> <Search class="absolute left-2.5 top-1/2 -translate-y-1/2 h-3.5 w-3.5 text-muted-foreground z-10 pointer-events-none" />
<Input <Input
id="usage-records-search" id="usage-records-search"
v-model="localSearch" v-model="localSearch"
:placeholder="isAdmin ? '搜索用户/密钥' : '搜索密钥/模型'" :placeholder="isAdmin ? '搜索用户/密钥' : '搜索密钥/模型'"
class="w-[7.5rem] sm:w-48 h-8 text-xs border-border/60 pl-8" class="w-[7.5rem] sm:w-48 h-8 text-xs border-border/60 pl-8"
/> />
</div> </div>
<!-- 用户筛选（仅管理员可见） --> <!-- 用户筛选（仅管理员可见） -->
<Select <Select
v-if="isAdmin && availableUsers.length > 0" v-if="isAdmin && availableUsers.length > 0"
:model-value="filterUser" :model-value="filterUser"
@update:model-value="$emit('update:filterUser', $event)" @update:model-value="$emit('update:filterUser', $event)"
>
<SelectTrigger class="flex-1 min-w-0 sm:flex-none sm:w-36 h-8 text-xs border-border/60">
<SelectValue placeholder="用户" />
</SelectTrigger>
<SelectContent>
<SelectItem value="__all__">
全部用户
</SelectItem>
<SelectItem
v-for="user in availableUsers"
:key="user.id"
:value="user.id"
> >
<SelectTrigger class="flex-1 min-w-0 sm:flex-none sm:w-36 h-8 text-xs border-border/60"> {{ user.username || user.email }}
<SelectValue placeholder="用户" /> </SelectItem>
</SelectTrigger> </SelectContent>
<SelectContent> </Select>
<SelectItem value="__all__">
全部用户
</SelectItem>
<SelectItem
v-for="user in availableUsers"
:key="user.id"
:value="user.id"
>
{{ user.username || user.email }}
</SelectItem>
</SelectContent>
</Select>
<!-- 模型筛选 --> <!-- 模型筛选 -->
<Select <Select
:model-value="filterModel" :model-value="filterModel"
@update:model-value="$emit('update:filterModel', $event)" @update:model-value="$emit('update:filterModel', $event)"
>
<SelectTrigger class="flex-1 min-w-0 sm:flex-none sm:w-40 h-8 text-xs border-border/60">
<SelectValue placeholder="模型" />
</SelectTrigger>
<SelectContent>
<SelectItem value="__all__">
全部模型
</SelectItem>
<SelectItem
v-for="model in availableModels"
:key="model"
:value="model"
> >
<SelectTrigger class="flex-1 min-w-0 sm:flex-none sm:w-40 h-8 text-xs border-border/60"> {{ model.replace('claude-', '') }}
<SelectValue placeholder="模型" /> </SelectItem>
</SelectTrigger> </SelectContent>
<SelectContent> </Select>
<SelectItem value="__all__">
全部模型
</SelectItem>
<SelectItem
v-for="model in availableModels"
:key="model"
:value="model"
>
{{ model.replace('claude-', '') }}
</SelectItem>
</SelectContent>
</Select>
<!-- 提供商筛选（仅管理员可见） --> <!-- 提供商筛选（仅管理员可见） -->
<Select <Select
v-if="isAdmin" v-if="isAdmin"
:model-value="filterProvider" :model-value="filterProvider"
@update:model-value="$emit('update:filterProvider', $event)" @update:model-value="$emit('update:filterProvider', $event)"
>
<SelectTrigger class="flex-1 min-w-0 sm:flex-none sm:w-32 h-8 text-xs border-border/60">
<SelectValue placeholder="提供商" />
</SelectTrigger>
<SelectContent>
<SelectItem value="__all__">
全部提供商
</SelectItem>
<SelectItem
v-for="provider in availableProviders"
:key="provider"
:value="provider"
> >
<SelectTrigger class="flex-1 min-w-0 sm:flex-none sm:w-32 h-8 text-xs border-border/60"> {{ provider }}
<SelectValue placeholder="提供商" /> </SelectItem>
</SelectTrigger> </SelectContent>
<SelectContent> </Select>
<SelectItem value="__all__">
全部提供商
</SelectItem>
<SelectItem
v-for="provider in availableProviders"
:key="provider"
:value="provider"
>
{{ provider }}
</SelectItem>
</SelectContent>
</Select>
<!-- API格式筛选 --> <!-- API格式筛选 -->
<Select <Select
:model-value="filterApiFormat" :model-value="filterApiFormat"
@update:model-value="$emit('update:filterApiFormat', $event)" @update:model-value="$emit('update:filterApiFormat', $event)"
>
<SelectTrigger class="flex-1 min-w-0 sm:flex-none sm:w-32 h-8 text-xs border-border/60">
<SelectValue placeholder="格式" />
</SelectTrigger>
<SelectContent>
<SelectItem value="__all__">
全部格式
</SelectItem>
<SelectItem
v-for="format in availableApiFormats"
:key="format.value"
:value="format.value"
> >
<SelectTrigger class="flex-1 min-w-0 sm:flex-none sm:w-32 h-8 text-xs border-border/60"> {{ format.label }}
<SelectValue placeholder="格式" /> </SelectItem>
</SelectTrigger> </SelectContent>
<SelectContent> </Select>
<SelectItem value="__all__">
全部格式
</SelectItem>
<SelectItem
v-for="format in availableApiFormats"
:key="format.value"
:value="format.value"
>
{{ format.label }}
</SelectItem>
</SelectContent>
</Select>
<!-- 状态筛选 --> <!-- 状态筛选 -->
<Select <Select
:model-value="filterStatus" :model-value="filterStatus"
@update:model-value="$emit('update:filterStatus', $event)" @update:model-value="$emit('update:filterStatus', $event)"
> >
<SelectTrigger class="flex-1 min-w-0 sm:flex-none sm:w-28 h-8 text-xs border-border/60"> <SelectTrigger class="flex-1 min-w-0 sm:flex-none sm:w-28 h-8 text-xs border-border/60">
<SelectValue placeholder="状态" /> <SelectValue placeholder="状态" />
</SelectTrigger> </SelectTrigger>
<SelectContent> <SelectContent>
<SelectItem value="__all__"> <SelectItem value="__all__">
全部状态 全部状态
</SelectItem> </SelectItem>
<SelectItem value="stream"> <SelectItem value="stream">
流式 流式
</SelectItem> </SelectItem>
<SelectItem value="standard"> <SelectItem value="standard">
标准 标准
</SelectItem> </SelectItem>
<SelectItem value="active"> <SelectItem value="active">
活跃 活跃
</SelectItem> </SelectItem>
<SelectItem value="failed"> <SelectItem value="failed">
失败 失败
</SelectItem> </SelectItem>
<SelectItem value="cancelled"> <SelectItem value="cancelled">
已取消 已取消
</SelectItem> </SelectItem>
<SelectItem value="has_retry"> <SelectItem value="has_retry">
发生重试 发生重试
</SelectItem> </SelectItem>
<SelectItem value="has_fallback"> <SelectItem value="has_fallback">
发生转移 发生转移
</SelectItem> </SelectItem>
</SelectContent> </SelectContent>
</Select> </Select>
<!-- 分隔线 --> <!-- 分隔线 -->
<div class="hidden sm:block h-4 w-px bg-border" /> <div class="hidden sm:block h-4 w-px bg-border" />
<!-- 自动刷新按钮 --> <!-- 自动刷新按钮 -->
<Button <Button
variant="ghost" variant="ghost"
size="icon" size="icon"
class="h-8 w-8" class="h-8 w-8"
:class="autoRefresh ? 'text-primary' : ''" :class="autoRefresh ? 'text-primary' : ''"
:title="autoRefresh ? '点击关闭自动刷新' : '点击开启自动刷新（每3秒刷新）'" :title="autoRefresh ? '点击关闭自动刷新' : '点击开启自动刷新（每3秒刷新）'"
@click="$emit('update:autoRefresh', !autoRefresh)" @click="$emit('update:autoRefresh', !autoRefresh)"
> >
<RefreshCcw <RefreshCcw
class="w-3.5 h-3.5" class="w-3.5 h-3.5"
:class="autoRefresh ? 'animate-spin' : ''" :class="autoRefresh ? 'animate-spin' : ''"
/> />
</Button> </Button>
</template> </template>
<!-- 移动端卡片视图 --> <!-- 移动端卡片视图 -->

18
frontend/src/features/usage/composables/useUsageData.ts
View File

@@ -110,7 +110,10 @@ export function useUsageData(options: UseUsageDataOptions) {
model: item.model, model: item.model,
request_count: item.request_count || 0, request_count: item.request_count || 0,
total_tokens: item.total_tokens || 0, total_tokens: item.total_tokens || 0,
total_input_context: typeof raw.total_input_context === 'number' ? raw.total_input_context : 0,
output_tokens: typeof raw.output_tokens === 'number' ? raw.output_tokens : 0,
cache_read_tokens: typeof raw.cache_read_tokens === 'number' ? raw.cache_read_tokens : 0, cache_read_tokens: typeof raw.cache_read_tokens === 'number' ? raw.cache_read_tokens : 0,
cache_creation_tokens: typeof raw.cache_creation_tokens === 'number' ? raw.cache_creation_tokens : 0,
cache_hit_rate: typeof raw.cache_hit_rate === 'number' ? raw.cache_hit_rate : 0, cache_hit_rate: typeof raw.cache_hit_rate === 'number' ? raw.cache_hit_rate : 0,
total_cost: item.total_cost || 0, total_cost: item.total_cost || 0,
actual_cost: typeof raw.actual_cost === 'number' ? raw.actual_cost : undefined actual_cost: typeof raw.actual_cost === 'number' ? raw.actual_cost : undefined
@@ -121,7 +124,10 @@ export function useUsageData(options: UseUsageDataOptions) {
provider: item.provider, provider: item.provider,
requests: item.request_count, requests: item.request_count,
totalTokens: item.total_tokens || 0, totalTokens: item.total_tokens || 0,
totalInputContext: item.total_input_context || 0,
outputTokens: item.output_tokens || 0,
cacheReadTokens: item.cache_read_tokens || 0, cacheReadTokens: item.cache_read_tokens || 0,
cacheCreationTokens: item.cache_creation_tokens || 0,
cacheHitRate: item.cache_hit_rate || 0, cacheHitRate: item.cache_hit_rate || 0,
totalCost: item.total_cost, totalCost: item.total_cost,
actualCost: item.actual_cost, actualCost: item.actual_cost,
@@ -135,7 +141,10 @@ export function useUsageData(options: UseUsageDataOptions) {
api_format: item.api_format, api_format: item.api_format,
request_count: item.request_count || 0, request_count: item.request_count || 0,
total_tokens: item.total_tokens || 0, total_tokens: item.total_tokens || 0,
total_input_context: item.total_input_context || 0,
output_tokens: item.output_tokens || 0,
cache_read_tokens: item.cache_read_tokens || 0, cache_read_tokens: item.cache_read_tokens || 0,
cache_creation_tokens: item.cache_creation_tokens || 0,
cache_hit_rate: item.cache_hit_rate || 0, cache_hit_rate: item.cache_hit_rate || 0,
total_cost: item.total_cost || 0, total_cost: item.total_cost || 0,
actual_cost: item.actual_cost, actual_cost: item.actual_cost,
@@ -169,7 +178,10 @@ export function useUsageData(options: UseUsageDataOptions) {
model: item.model, model: item.model,
request_count: item.requests || 0, request_count: item.requests || 0,
total_tokens: item.total_tokens || 0, total_tokens: item.total_tokens || 0,
total_input_context: item.total_input_context || 0,
output_tokens: item.output_tokens || 0,
cache_read_tokens: item.cache_read_tokens || 0, cache_read_tokens: item.cache_read_tokens || 0,
cache_creation_tokens: item.cache_creation_tokens || 0,
cache_hit_rate: item.cache_hit_rate || 0, cache_hit_rate: item.cache_hit_rate || 0,
total_cost: item.total_cost_usd || 0, total_cost: item.total_cost_usd || 0,
actual_cost: item.actual_total_cost_usd actual_cost: item.actual_total_cost_usd
@@ -179,7 +191,10 @@ export function useUsageData(options: UseUsageDataOptions) {
provider: item.provider, provider: item.provider,
requests: item.requests || 0, requests: item.requests || 0,
totalTokens: item.total_tokens || 0, totalTokens: item.total_tokens || 0,
totalInputContext: item.total_input_context || 0,
outputTokens: item.output_tokens || 0,
cacheReadTokens: item.cache_read_tokens || 0, cacheReadTokens: item.cache_read_tokens || 0,
cacheCreationTokens: item.cache_creation_tokens || 0,
cacheHitRate: item.cache_hit_rate || 0, cacheHitRate: item.cache_hit_rate || 0,
totalCost: item.total_cost_usd || 0, totalCost: item.total_cost_usd || 0,
successRate: item.success_rate || 0, successRate: item.success_rate || 0,
@@ -209,7 +224,10 @@ export function useUsageData(options: UseUsageDataOptions) {
api_format: item.api_format, api_format: item.api_format,
request_count: item.request_count || 0, request_count: item.request_count || 0,
total_tokens: item.total_tokens || 0, total_tokens: item.total_tokens || 0,
total_input_context: item.total_input_context || 0,
output_tokens: item.output_tokens || 0,
cache_read_tokens: item.cache_read_tokens || 0, cache_read_tokens: item.cache_read_tokens || 0,
cache_creation_tokens: item.cache_creation_tokens || 0,
cache_hit_rate: item.cache_hit_rate || 0, cache_hit_rate: item.cache_hit_rate || 0,
total_cost: item.total_cost_usd || 0, total_cost: item.total_cost_usd || 0,
avgResponseTime: (item.avg_response_time_ms ?? 0) > 0 avgResponseTime: (item.avg_response_time_ms ?? 0) > 0

9
frontend/src/features/usage/types.ts
View File

@@ -22,7 +22,10 @@ export interface ModelStatsItem {
model: string model: string
request_count: number request_count: number
total_tokens: number total_tokens: number
total_input_context?: number
output_tokens?: number
cache_read_tokens?: number cache_read_tokens?: number
cache_creation_tokens?: number
cache_hit_rate?: number cache_hit_rate?: number
total_cost: number total_cost: number
actual_cost?: number // 倍率消耗 actual_cost?: number // 倍率消耗
@@ -38,7 +41,10 @@ export interface ProviderStatsItem {
provider: string provider: string
requests: number requests: number
totalTokens: number totalTokens: number
totalInputContext?: number
outputTokens?: number
cacheReadTokens?: number cacheReadTokens?: number
cacheCreationTokens?: number
cacheHitRate?: number cacheHitRate?: number
totalCost: number totalCost: number
actualCost?: number actualCost?: number
@@ -51,7 +57,10 @@ export interface ApiFormatStatsItem {
api_format: string api_format: string
request_count: number request_count: number
total_tokens: number total_tokens: number
total_input_context?: number
output_tokens?: number
cache_read_tokens?: number cache_read_tokens?: number
cache_creation_tokens?: number
cache_hit_rate?: number cache_hit_rate?: number
total_cost: number total_cost: number
actual_cost?: number actual_cost?: number

5
frontend/src/layouts/MainLayout.vue
View File

@@ -282,7 +282,10 @@
<div class="flex items-center gap-2"> <div class="flex items-center gap-2">
<!-- Page-level header actions (right side) --> <!-- Page-level header actions (right side) -->
<div id="header-actions-right" class="flex items-center" /> <div
id="header-actions-right"
class="flex items-center"
/>
<!-- Theme Toggle --> <!-- Theme Toggle -->
<button <button
class="flex h-9 w-9 items-center justify-center rounded-lg text-muted-foreground hover:text-foreground hover:bg-muted/50 transition" class="flex h-9 w-9 items-center justify-center rounded-lg text-muted-foreground hover:text-foreground hover:bg-muted/50 transition"

5
frontend/src/views/shared/Usage.vue
View File

@@ -1,7 +1,10 @@
<template> <template>
<div class="space-y-6 pb-8"> <div class="space-y-6 pb-8">
<!-- 面包屑旁的折叠按钮 --> <!-- 面包屑旁的折叠按钮 -->
<Teleport to="#header-actions-right" defer> <Teleport
to="#header-actions-right"
defer
>
<button <button
class="flex h-9 w-9 items-center justify-center rounded-lg text-muted-foreground hover:text-foreground hover:bg-muted/50 transition" class="flex h-9 w-9 items-center justify-center rounded-lg text-muted-foreground hover:text-foreground hover:bg-muted/50 transition"
:title="statsExpanded ? '收起用量分析' : '展开用量分析'" :title="statsExpanded ? '收起用量分析' : '展开用量分析'"

33
src/api/admin/usage/routes.py
View File

@@ -546,6 +546,8 @@ class AdminUsageByModelAdapter(AdminApiAdapter):
func.sum(Usage.actual_total_cost_usd).label("actual_cost"), func.sum(Usage.actual_total_cost_usd).label("actual_cost"),
func.sum(Usage.cache_read_input_tokens).label("cache_read_tokens"), func.sum(Usage.cache_read_input_tokens).label("cache_read_tokens"),
func.sum(input_context_expr()).label("total_input_context"), func.sum(input_context_expr()).label("total_input_context"),
func.sum(Usage.output_tokens).label("output_tokens"),
func.sum(Usage.cache_creation_input_tokens).label("cache_creation_tokens"),
) )
# 过滤掉 pending/streaming 状态的请求（尚未完成的请求不应计入统计） # 过滤掉 pending/streaming 状态的请求（尚未完成的请求不应计入统计）
query = query.filter(Usage.status.notin_(["pending", "streaming"])) query = query.filter(Usage.status.notin_(["pending", "streaming"]))
@@ -573,15 +575,28 @@ class AdminUsageByModelAdapter(AdminApiAdapter):
"model": model, "model": model,
"request_count": count, "request_count": count,
"total_tokens": int(tokens or 0), "total_tokens": int(tokens or 0),
"total_input_context": int(total_input_context or 0),
"output_tokens": int(output_tokens or 0),
"total_cost": float(cost or 0), "total_cost": float(cost or 0),
"actual_cost": float(actual_cost or 0), "actual_cost": float(actual_cost or 0),
"cache_read_tokens": int(cache_read_tokens or 0), "cache_read_tokens": int(cache_read_tokens or 0),
"cache_creation_tokens": int(cache_creation_tokens or 0),
"cache_hit_rate": _calculate_token_cache_hit_rate( "cache_hit_rate": _calculate_token_cache_hit_rate(
total_input_context=total_input_context, total_input_context=total_input_context,
cache_read_tokens=cache_read_tokens, cache_read_tokens=cache_read_tokens,
), ),
} }
for model, count, tokens, cost, actual_cost, cache_read_tokens, total_input_context in stats for (
model,
count,
tokens,
cost,
actual_cost,
cache_read_tokens,
total_input_context,
output_tokens,
cache_creation_tokens,
) in stats
] ]
@@ -707,6 +722,8 @@ class AdminUsageByProviderAdapter(AdminApiAdapter):
func.avg(Usage.response_time_ms).label("avg_response_time_ms"), func.avg(Usage.response_time_ms).label("avg_response_time_ms"),
func.sum(Usage.cache_read_input_tokens).label("cache_read_tokens"), func.sum(Usage.cache_read_input_tokens).label("cache_read_tokens"),
func.sum(input_context_expr()).label("total_input_context"), func.sum(input_context_expr()).label("total_input_context"),
func.sum(Usage.output_tokens).label("output_tokens"),
func.sum(Usage.cache_creation_input_tokens).label("cache_creation_tokens"),
).filter( ).filter(
Usage.provider_id.isnot(None), Usage.provider_id.isnot(None),
# 过滤掉 pending/streaming 状态的请求 # 过滤掉 pending/streaming 状态的请求
@@ -766,6 +783,10 @@ class AdminUsageByProviderAdapter(AdminApiAdapter):
"provider": provider_map.get(provider_id_str, "Unknown"), "provider": provider_map.get(provider_id_str, "Unknown"),
"request_count": attempt_count, # 尝试次数 "request_count": attempt_count, # 尝试次数
"total_tokens": int(usage_stat.total_tokens or 0) if usage_stat else 0, "total_tokens": int(usage_stat.total_tokens or 0) if usage_stat else 0,
"total_input_context": (
int(usage_stat.total_input_context or 0) if usage_stat else 0
),
"output_tokens": int(usage_stat.output_tokens or 0) if usage_stat else 0,
"total_cost": float(usage_stat.total_cost or 0) if usage_stat else 0, "total_cost": float(usage_stat.total_cost or 0) if usage_stat else 0,
"actual_cost": float(usage_stat.actual_cost or 0) if usage_stat else 0, "actual_cost": float(usage_stat.actual_cost or 0) if usage_stat else 0,
"avg_response_time_ms": float(stat.avg_latency_ms or 0), "avg_response_time_ms": float(stat.avg_latency_ms or 0),
@@ -774,6 +795,9 @@ class AdminUsageByProviderAdapter(AdminApiAdapter):
"cache_read_tokens": ( "cache_read_tokens": (
int(usage_stat.cache_read_tokens or 0) if usage_stat else 0 int(usage_stat.cache_read_tokens or 0) if usage_stat else 0
), ),
"cache_creation_tokens": (
int(usage_stat.cache_creation_tokens or 0) if usage_stat else 0
),
"cache_hit_rate": _calculate_token_cache_hit_rate( "cache_hit_rate": _calculate_token_cache_hit_rate(
total_input_context=(usage_stat.total_input_context if usage_stat else 0), total_input_context=(usage_stat.total_input_context if usage_stat else 0),
cache_read_tokens=(usage_stat.cache_read_tokens if usage_stat else 0), cache_read_tokens=(usage_stat.cache_read_tokens if usage_stat else 0),
@@ -811,6 +835,8 @@ class AdminUsageByApiFormatAdapter(AdminApiAdapter):
func.avg(Usage.response_time_ms).label("avg_response_time_ms"), func.avg(Usage.response_time_ms).label("avg_response_time_ms"),
func.sum(Usage.cache_read_input_tokens).label("cache_read_tokens"), func.sum(Usage.cache_read_input_tokens).label("cache_read_tokens"),
func.sum(input_context_expr()).label("total_input_context"), func.sum(input_context_expr()).label("total_input_context"),
func.sum(Usage.output_tokens).label("output_tokens"),
func.sum(Usage.cache_creation_input_tokens).label("cache_creation_tokens"),
) )
# 过滤掉 pending/streaming 状态的请求 # 过滤掉 pending/streaming 状态的请求
query = query.filter(Usage.status.notin_(["pending", "streaming"])) query = query.filter(Usage.status.notin_(["pending", "streaming"]))
@@ -843,10 +869,13 @@ class AdminUsageByApiFormatAdapter(AdminApiAdapter):
"api_format": api_format or "unknown", "api_format": api_format or "unknown",
"request_count": count, "request_count": count,
"total_tokens": int(tokens or 0), "total_tokens": int(tokens or 0),
"total_input_context": int(total_input_context or 0),
"output_tokens": int(output_tokens or 0),
"total_cost": float(cost or 0), "total_cost": float(cost or 0),
"actual_cost": float(actual_cost or 0), "actual_cost": float(actual_cost or 0),
"avg_response_time_ms": float(avg_response_time or 0), "avg_response_time_ms": float(avg_response_time or 0),
"cache_read_tokens": int(cache_read_tokens or 0), "cache_read_tokens": int(cache_read_tokens or 0),
"cache_creation_tokens": int(cache_creation_tokens or 0),
"cache_hit_rate": _calculate_token_cache_hit_rate( "cache_hit_rate": _calculate_token_cache_hit_rate(
total_input_context=total_input_context, total_input_context=total_input_context,
cache_read_tokens=cache_read_tokens, cache_read_tokens=cache_read_tokens,
@@ -861,6 +890,8 @@ class AdminUsageByApiFormatAdapter(AdminApiAdapter):
avg_response_time, avg_response_time,
cache_read_tokens, cache_read_tokens,
total_input_context, total_input_context,
output_tokens,
cache_creation_tokens,
) in stats ) in stats
] ]

19
src/api/user_me/routes.py
View File

@@ -1061,6 +1061,7 @@ class GetUsageAdapter(AuthenticatedApiAdapter):
"output_tokens": 0, "output_tokens": 0,
"total_tokens": 0, "total_tokens": 0,
"cache_read_tokens": 0, "cache_read_tokens": 0,
"cache_creation_tokens": 0,
"total_input_context": 0, "total_input_context": 0,
"cache_hit_rate": 0.0, "cache_hit_rate": 0.0,
"total_cost_usd": 0.0, "total_cost_usd": 0.0,
@@ -1075,6 +1076,7 @@ class GetUsageAdapter(AuthenticatedApiAdapter):
stats["output_tokens"] += item["output_tokens"] stats["output_tokens"] += item["output_tokens"]
stats["total_tokens"] += item["total_tokens"] stats["total_tokens"] += item["total_tokens"]
stats["cache_read_tokens"] += int(item.get("cache_read_tokens", 0) or 0) stats["cache_read_tokens"] += int(item.get("cache_read_tokens", 0) or 0)
stats["cache_creation_tokens"] += int(item.get("cache_creation_tokens", 0) or 0)
stats["total_input_context"] += int(item.get("total_input_context", 0) or 0) stats["total_input_context"] += int(item.get("total_input_context", 0) or 0)
stats["total_cost_usd"] += item["total_cost_usd"] stats["total_cost_usd"] += item["total_cost_usd"]
# 管理员可以看到真实成本 # 管理员可以看到真实成本
@@ -1086,7 +1088,9 @@ class GetUsageAdapter(AuthenticatedApiAdapter):
"provider": provider_name, "provider": provider_name,
"requests": 0, "requests": 0,
"total_tokens": 0, "total_tokens": 0,
"output_tokens": 0,
"cache_read_tokens": 0, "cache_read_tokens": 0,
"cache_creation_tokens": 0,
"total_input_context": 0, "total_input_context": 0,
"total_cost_usd": 0.0, "total_cost_usd": 0.0,
"success_count": 0, "success_count": 0,
@@ -1096,7 +1100,11 @@ class GetUsageAdapter(AuthenticatedApiAdapter):
provider_stats = provider_summary.setdefault(provider_name, provider_base_stats) provider_stats = provider_summary.setdefault(provider_name, provider_base_stats)
provider_stats["requests"] += item["requests"] provider_stats["requests"] += item["requests"]
provider_stats["total_tokens"] += item["total_tokens"] provider_stats["total_tokens"] += item["total_tokens"]
provider_stats["output_tokens"] += item.get("output_tokens", 0) or 0
provider_stats["cache_read_tokens"] += int(item.get("cache_read_tokens", 0) or 0) provider_stats["cache_read_tokens"] += int(item.get("cache_read_tokens", 0) or 0)
provider_stats["cache_creation_tokens"] += int(
item.get("cache_creation_tokens", 0) or 0
)
provider_stats["total_input_context"] += int(item.get("total_input_context", 0) or 0) provider_stats["total_input_context"] += int(item.get("total_input_context", 0) or 0)
provider_stats["total_cost_usd"] += item["total_cost_usd"] provider_stats["total_cost_usd"] += item["total_cost_usd"]
provider_stats["success_count"] += int(item.get("success_count", 0) or 0) provider_stats["success_count"] += int(item.get("success_count", 0) or 0)
@@ -1112,7 +1120,6 @@ class GetUsageAdapter(AuthenticatedApiAdapter):
total_input_context=int(model_stats.get("total_input_context", 0) or 0), total_input_context=int(model_stats.get("total_input_context", 0) or 0),
cache_read_tokens=int(model_stats.get("cache_read_tokens", 0) or 0), cache_read_tokens=int(model_stats.get("cache_read_tokens", 0) or 0),
) )
model_stats.pop("total_input_context", None)
summary_by_model = sorted(model_summary.values(), key=lambda x: x["requests"], reverse=True) summary_by_model = sorted(model_summary.values(), key=lambda x: x["requests"], reverse=True)
summary_by_provider = [] summary_by_provider = []
@@ -1132,7 +1139,10 @@ class GetUsageAdapter(AuthenticatedApiAdapter):
"provider": provider_stats["provider"], "provider": provider_stats["provider"],
"requests": provider_stats["requests"], "requests": provider_stats["requests"],
"total_tokens": provider_stats["total_tokens"], "total_tokens": provider_stats["total_tokens"],
"total_input_context": provider_stats["total_input_context"],
"output_tokens": provider_stats["output_tokens"],
"cache_read_tokens": provider_stats["cache_read_tokens"], "cache_read_tokens": provider_stats["cache_read_tokens"],
"cache_creation_tokens": provider_stats["cache_creation_tokens"],
"cache_hit_rate": _calculate_token_cache_hit_rate( "cache_hit_rate": _calculate_token_cache_hit_rate(
total_input_context=int(provider_stats.get("total_input_context", 0) or 0), total_input_context=int(provider_stats.get("total_input_context", 0) or 0),
cache_read_tokens=int(provider_stats.get("cache_read_tokens", 0) or 0), cache_read_tokens=int(provider_stats.get("cache_read_tokens", 0) or 0),
@@ -1151,6 +1161,8 @@ class GetUsageAdapter(AuthenticatedApiAdapter):
func.sum(Usage.total_tokens).label("total_tokens"), func.sum(Usage.total_tokens).label("total_tokens"),
func.sum(Usage.cache_read_input_tokens).label("cache_read_tokens"), func.sum(Usage.cache_read_input_tokens).label("cache_read_tokens"),
func.sum(input_context_expr()).label("total_input_context"), func.sum(input_context_expr()).label("total_input_context"),
func.sum(Usage.output_tokens).label("output_tokens"),
func.sum(Usage.cache_creation_input_tokens).label("cache_creation_tokens"),
func.sum(Usage.total_cost_usd).label("total_cost_usd"), func.sum(Usage.total_cost_usd).label("total_cost_usd"),
func.avg(Usage.response_time_ms).label("avg_response_time_ms"), func.avg(Usage.response_time_ms).label("avg_response_time_ms"),
).filter( ).filter(
@@ -1171,7 +1183,10 @@ class GetUsageAdapter(AuthenticatedApiAdapter):
"api_format": api_format or "unknown", "api_format": api_format or "unknown",
"request_count": count, "request_count": count,
"total_tokens": int(total_tokens or 0), "total_tokens": int(total_tokens or 0),
"total_input_context": int(total_input_context or 0),
"output_tokens": int(output_tokens or 0),
"cache_read_tokens": int(cache_read_tokens or 0), "cache_read_tokens": int(cache_read_tokens or 0),
"cache_creation_tokens": int(cache_creation_tokens or 0),
"cache_hit_rate": _calculate_token_cache_hit_rate( "cache_hit_rate": _calculate_token_cache_hit_rate(
total_input_context=total_input_context, total_input_context=total_input_context,
cache_read_tokens=cache_read_tokens, cache_read_tokens=cache_read_tokens,
@@ -1185,6 +1200,8 @@ class GetUsageAdapter(AuthenticatedApiAdapter):
total_tokens, total_tokens,
cache_read_tokens, cache_read_tokens,
total_input_context, total_input_context,
output_tokens,
cache_creation_tokens,
total_cost_usd, total_cost_usd,
avg_response_time_ms, avg_response_time_ms,
) in api_format_stats ) in api_format_stats

2
src/services/usage/query.py
View File

@@ -261,6 +261,7 @@ class UsageQueryMixin:
func.sum(Usage.output_tokens).label("output_tokens"), func.sum(Usage.output_tokens).label("output_tokens"),
func.sum(Usage.total_tokens).label("total_tokens"), func.sum(Usage.total_tokens).label("total_tokens"),
func.sum(Usage.cache_read_input_tokens).label("cache_read_tokens"), func.sum(Usage.cache_read_input_tokens).label("cache_read_tokens"),
func.sum(Usage.cache_creation_input_tokens).label("cache_creation_tokens"),
func.sum(input_context_expr()).label("total_input_context"), func.sum(input_context_expr()).label("total_input_context"),
func.sum(Usage.total_cost_usd).label("total_cost_usd"), func.sum(Usage.total_cost_usd).label("total_cost_usd"),
func.sum(Usage.actual_total_cost_usd).label("actual_total_cost_usd"), func.sum(Usage.actual_total_cost_usd).label("actual_total_cost_usd"),
@@ -307,6 +308,7 @@ class UsageQueryMixin:
"output_tokens": row.output_tokens, "output_tokens": row.output_tokens,
"total_tokens": row.total_tokens, "total_tokens": row.total_tokens,
"cache_read_tokens": int(row.cache_read_tokens or 0), "cache_read_tokens": int(row.cache_read_tokens or 0),
"cache_creation_tokens": int(row.cache_creation_tokens or 0),
"total_input_context": int(row.total_input_context or 0), "total_input_context": int(row.total_input_context or 0),
"total_cost_usd": float(row.total_cost_usd or 0.0), "total_cost_usd": float(row.total_cost_usd or 0.0),
"actual_total_cost_usd": float(row.actual_total_cost_usd or 0.0), "actual_total_cost_usd": float(row.actual_total_cost_usd or 0.0),