Show raw input token counts in json output. (#15021)

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

packages/cli/src/ui/components/ModelStatsDisplay.test.tsx

@@ -87,6 +87,7 @@ describe('<ModelStatsDisplay />', () => {
         'gemini-2.5-pro': {
           api: { totalRequests: 1, totalErrors: 0, totalLatencyMs: 100 },
           tokens: {
+            input: 10,
             prompt: 10,
             candidates: 20,
             total: 30,
@@ -128,6 +129,7 @@ describe('<ModelStatsDisplay />', () => {
         'gemini-2.5-pro': {
           api: { totalRequests: 1, totalErrors: 0, totalLatencyMs: 100 },
           tokens: {
+            input: 5,
             prompt: 10,
             candidates: 20,
             total: 30,
@@ -139,6 +141,7 @@ describe('<ModelStatsDisplay />', () => {
         'gemini-2.5-flash': {
           api: { totalRequests: 1, totalErrors: 0, totalLatencyMs: 50 },
           tokens: {
+            input: 5,
             prompt: 5,
             candidates: 10,
             total: 15,
@@ -180,6 +183,7 @@ describe('<ModelStatsDisplay />', () => {
         'gemini-2.5-pro': {
           api: { totalRequests: 10, totalErrors: 1, totalLatencyMs: 1000 },
           tokens: {
+            input: 50,
             prompt: 100,
             candidates: 200,
             total: 300,
@@ -191,6 +195,7 @@ describe('<ModelStatsDisplay />', () => {
         'gemini-2.5-flash': {
           api: { totalRequests: 20, totalErrors: 2, totalLatencyMs: 500 },
           tokens: {
+            input: 100,
             prompt: 200,
             candidates: 400,
             total: 600,
@@ -235,6 +240,7 @@ describe('<ModelStatsDisplay />', () => {
             totalLatencyMs: 9876,
           },
           tokens: {
+            input: 987654321 - 123456789,
             prompt: 987654321,
             candidates: 123456789,
             total: 999999999,
@@ -272,6 +278,7 @@ describe('<ModelStatsDisplay />', () => {
         'gemini-2.5-pro': {
           api: { totalRequests: 1, totalErrors: 0, totalLatencyMs: 100 },
           tokens: {
+            input: 5,
             prompt: 10,
             candidates: 20,
             total: 30,

packages/cli/src/ui/components/ModelStatsDisplay.tsx

@@ -170,7 +170,7 @@ export const ModelStatsDisplay: React.FC = () => {
         isSubtle
         values={getModelValues((m) => (
           <Text color={theme.text.primary}>
-            {Math.max(0, m.tokens.prompt - m.tokens.cached).toLocaleString()}
+            {m.tokens.input.toLocaleString()}
           </Text>
         ))}
       />

packages/cli/src/ui/components/SessionSummaryDisplay.test.tsx

@@ -45,6 +45,7 @@ describe('<SessionSummaryDisplay />', () => {
         'gemini-2.5-pro': {
           api: { totalRequests: 10, totalErrors: 1, totalLatencyMs: 50234 },
           tokens: {
+            input: 500,
             prompt: 1000,
             candidates: 2000,
             total: 3500,

packages/cli/src/ui/components/StatsDisplay.test.tsx

@@ -85,6 +85,7 @@ describe('<StatsDisplay />', () => {
         'gemini-2.5-pro': {
           api: { totalRequests: 3, totalErrors: 0, totalLatencyMs: 15000 },
           tokens: {
+            input: 500,
             prompt: 1000,
             candidates: 2000,
             total: 43234,
@@ -96,6 +97,7 @@ describe('<StatsDisplay />', () => {
         'gemini-2.5-flash': {
           api: { totalRequests: 5, totalErrors: 1, totalLatencyMs: 4500 },
           tokens: {
+            input: 15000,
             prompt: 25000,
             candidates: 15000,
             total: 150000000,
@@ -123,6 +125,7 @@ describe('<StatsDisplay />', () => {
         'gemini-2.5-pro': {
           api: { totalRequests: 1, totalErrors: 0, totalLatencyMs: 100 },
           tokens: {
+            input: 50,
             prompt: 100,
             candidates: 100,
             total: 250,
@@ -216,6 +219,7 @@ describe('<StatsDisplay />', () => {
           'gemini-2.5-pro': {
             api: { totalRequests: 1, totalErrors: 0, totalLatencyMs: 100 },
             tokens: {
+              input: 100,
               prompt: 100,
               candidates: 100,
               total: 200,
@@ -398,6 +402,7 @@ describe('<StatsDisplay />', () => {
           'gemini-2.5-pro': {
             api: { totalRequests: 1, totalErrors: 0, totalLatencyMs: 100 },
             tokens: {
+              input: 50,
               prompt: 100,
               candidates: 100,
               total: 250,

packages/cli/src/ui/components/StatsDisplay.tsx

@@ -85,15 +85,13 @@ const buildModelRows = (
   const activeRows = Object.entries(models).map(([name, metrics]) => {
     const modelName = getBaseModelName(name);
     const cachedTokens = metrics.tokens.cached;
-    const totalInputTokens = metrics.tokens.prompt;
-    const uncachedTokens = Math.max(0, totalInputTokens - cachedTokens);
+    const inputTokens = metrics.tokens.input;
     return {
       key: name,
       modelName,
       requests: metrics.api.totalRequests,
       cachedTokens: cachedTokens.toLocaleString(),
-      uncachedTokens: uncachedTokens.toLocaleString(),
-      totalInputTokens: totalInputTokens.toLocaleString(),
+      inputTokens: inputTokens.toLocaleString(),
       outputTokens: metrics.tokens.candidates.toLocaleString(),
       bucket: quotas?.buckets?.find((b) => b.modelId === modelName),
       isActive: true,
@@ -114,8 +112,7 @@ const buildModelRows = (
         modelName: bucket.modelId!,
         requests: '-',
         cachedTokens: '-',
-        uncachedTokens: '-',
-        totalInputTokens: '-',
+        inputTokens: '-',
         outputTokens: '-',
         bucket,
         isActive: false,
@@ -290,7 +287,7 @@ const ModelUsageTable: React.FC<{
                     row.isActive ? theme.text.primary : theme.text.secondary
                   }
                 >
-                  {row.uncachedTokens}
+                  {row.inputTokens}
                 </Text>
               </Box>
               <Box