MediaMetz/gemini-cli
1
0
Fork 0
mirror of https://github.com/google-gemini/gemini-cli.git synced 2026-04-24 20:14:44 -07:00
Code Issues Packages Projects Releases Wiki Activity

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

Browse Source 
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
Jacob Richman
2025-12-15 18:47:39 -08:00
committed by GitHub
parent bb0c0d8ee3
commit 79f664d593
17 changed files with 189 additions and 129 deletions
Download Patch File Download Diff File Expand all files Collapse all files
packages/core/src/output/json-formatter.test.ts
+2
View File
@@ -71,6 +71,7 @@ describe('JsonFormatter', () => {
totalLatencyMs: 5672,
},
tokens: {
input: 13745,
prompt: 24401,
candidates: 215,
total: 24719,
@@ -86,6 +87,7 @@ describe('JsonFormatter', () => {
totalLatencyMs: 5914,
},
tokens: {
input: 20803,
prompt: 20803,
candidates: 716,
total: 21657,
packages/core/src/output/stream-json-formatter.test.ts
+119 -117
View File
@@ -150,6 +150,8 @@ describe('StreamJsonFormatter', () => {
total_tokens: 100,
input_tokens: 50,
output_tokens: 50,
cached: 0,
input: 50,
duration_ms: 1200,
tool_calls: 2,
},
@@ -174,6 +176,8 @@ describe('StreamJsonFormatter', () => {
total_tokens: 100,
input_tokens: 50,
output_tokens: 50,
cached: 0,
input: 50,
duration_ms: 1200,
tool_calls: 0,
},
@@ -247,41 +251,114 @@ describe('StreamJsonFormatter', () => {
});
describe('convertToStreamStats', () => {
const createMockMetrics = (): SessionMetrics => ({
models: {},
tools: {
totalCalls: 0,
totalSuccess: 0,
totalFail: 0,
totalDurationMs: 0,
totalDecisions: {
[ToolCallDecision.ACCEPT]: 0,
[ToolCallDecision.REJECT]: 0,
[ToolCallDecision.MODIFY]: 0,
[ToolCallDecision.AUTO_ACCEPT]: 0,
},
byName: {},
},
files: {
totalLinesAdded: 0,
totalLinesRemoved: 0,
},
});
it('should aggregate token counts from single model', () => {
const metrics: SessionMetrics = {
models: {
'gemini-2.0-flash': {
api: {
totalRequests: 1,
totalErrors: 0,
totalLatencyMs: 1000,
},
tokens: {
prompt: 50,
candidates: 30,
total: 80,
cached: 0,
thoughts: 0,
tool: 0,
},
},
const metrics = createMockMetrics();
metrics.models['gemini-2.0-flash'] = {
api: {
totalRequests: 1,
totalErrors: 0,
totalLatencyMs: 1000,
},
tools: {
totalCalls: 2,
totalSuccess: 2,
totalFail: 0,
totalDurationMs: 500,
totalDecisions: {
[ToolCallDecision.ACCEPT]: 0,
[ToolCallDecision.REJECT]: 0,
[ToolCallDecision.MODIFY]: 0,
[ToolCallDecision.AUTO_ACCEPT]: 2,
},
byName: {},
tokens: {
input: 50,
prompt: 50,
candidates: 30,
total: 80,
cached: 0,
thoughts: 0,
tool: 0,
},
files: {
totalLinesAdded: 0,
totalLinesRemoved: 0,
};
metrics.tools.totalCalls = 2;
metrics.tools.totalDecisions[ToolCallDecision.AUTO_ACCEPT] = 2;
const result = formatter.convertToStreamStats(metrics, 1200);
expect(result).toEqual({
total_tokens: 80,
input_tokens: 50,
output_tokens: 30,
cached: 0,
input: 50,
duration_ms: 1200,
tool_calls: 2,
});
});
it('should aggregate token counts from multiple models', () => {
const metrics = createMockMetrics();
metrics.models['gemini-pro'] = {
api: { totalRequests: 1, totalErrors: 0, totalLatencyMs: 1000 },
tokens: {
input: 50,
prompt: 50,
candidates: 30,
total: 80,
cached: 0,
thoughts: 0,
tool: 0,
},
};
metrics.models['gemini-ultra'] = {
api: { totalRequests: 1, totalErrors: 0, totalLatencyMs: 2000 },
tokens: {
input: 100,
prompt: 100,
candidates: 70,
total: 170,
cached: 0,
thoughts: 0,
tool: 0,
},
};
metrics.tools.totalCalls = 5;
const result = formatter.convertToStreamStats(metrics, 3000);
expect(result).toEqual({
total_tokens: 250, // 80 + 170
input_tokens: 150, // 50 + 100
output_tokens: 100, // 30 + 70
cached: 0,
input: 150,
duration_ms: 3000,
tool_calls: 5,
});
});
it('should aggregate cached token counts correctly', () => {
const metrics = createMockMetrics();
metrics.models['gemini-pro'] = {
api: { totalRequests: 1, totalErrors: 0, totalLatencyMs: 1000 },
tokens: {
input: 20, // 50 prompt - 30 cached
prompt: 50,
candidates: 30,
total: 80,
cached: 30,
thoughts: 0,
tool: 0,
},
};
@@ -291,96 +368,15 @@ describe('StreamJsonFormatter', () => {
total_tokens: 80,
input_tokens: 50,
output_tokens: 30,
cached: 30,
input: 20,
duration_ms: 1200,
tool_calls: 2,
});
});
it('should aggregate token counts from multiple models', () => {
const metrics: SessionMetrics = {
models: {
'gemini-2.0-flash': {
api: {
totalRequests: 1,
totalErrors: 0,
totalLatencyMs: 1000,
},
tokens: {
prompt: 50,
candidates: 30,
total: 80,
cached: 0,
thoughts: 0,
tool: 0,
},
},
'gemini-1.5-pro': {
api: {
totalRequests: 1,
totalErrors: 0,
totalLatencyMs: 2000,
},
tokens: {
prompt: 100,
candidates: 70,
total: 170,
cached: 0,
thoughts: 0,
tool: 0,
},
},
},
tools: {
totalCalls: 5,
totalSuccess: 5,
totalFail: 0,
totalDurationMs: 1000,
totalDecisions: {
[ToolCallDecision.ACCEPT]: 0,
[ToolCallDecision.REJECT]: 0,
[ToolCallDecision.MODIFY]: 0,
[ToolCallDecision.AUTO_ACCEPT]: 5,
},
byName: {},
},
files: {
totalLinesAdded: 0,
totalLinesRemoved: 0,
},
};
const result = formatter.convertToStreamStats(metrics, 3000);
expect(result).toEqual({
total_tokens: 250, // 80 + 170
input_tokens: 150, // 50 + 100
output_tokens: 100, // 30 + 70
duration_ms: 3000,
tool_calls: 5,
tool_calls: 0,
});
});
it('should handle empty metrics', () => {
const metrics: SessionMetrics = {
models: {},
tools: {
totalCalls: 0,
totalSuccess: 0,
totalFail: 0,
totalDurationMs: 0,
totalDecisions: {
[ToolCallDecision.ACCEPT]: 0,
[ToolCallDecision.REJECT]: 0,
[ToolCallDecision.MODIFY]: 0,
[ToolCallDecision.AUTO_ACCEPT]: 0,
},
byName: {},
},
files: {
totalLinesAdded: 0,
totalLinesRemoved: 0,
},
};
const metrics = createMockMetrics();
const result = formatter.convertToStreamStats(metrics, 100);
@@ -388,6 +384,8 @@ describe('StreamJsonFormatter', () => {
total_tokens: 0,
input_tokens: 0,
output_tokens: 0,
cached: 0,
input: 0,
duration_ms: 100,
tool_calls: 0,
});
@@ -515,6 +513,8 @@ describe('StreamJsonFormatter', () => {
total_tokens: 0,
input_tokens: 0,
output_tokens: 0,
cached: 0,
input: 0,
duration_ms: 0,
tool_calls: 0,
},
@@ -536,6 +536,8 @@ describe('StreamJsonFormatter', () => {
total_tokens: 100,
input_tokens: 50,
output_tokens: 50,
cached: 0,
input: 50,
duration_ms: 1200,
tool_calls: 2,
},
packages/core/src/output/stream-json-formatter.ts
+6
View File
@@ -43,18 +43,24 @@ export class StreamJsonFormatter {
let totalTokens = 0;
let inputTokens = 0;
let outputTokens = 0;
let cached = 0;
let input = 0;
// Aggregate token counts across all models
for (const modelMetrics of Object.values(metrics.models)) {
totalTokens += modelMetrics.tokens.total;
inputTokens += modelMetrics.tokens.prompt;
outputTokens += modelMetrics.tokens.candidates;
cached += modelMetrics.tokens.cached;
input += modelMetrics.tokens.input;
}
return {
total_tokens: totalTokens,
input_tokens: inputTokens,
output_tokens: outputTokens,
cached,
input,
duration_ms: durationMs,
tool_calls: metrics.tools.totalCalls,
};
packages/core/src/output/types.ts
+3
View File
@@ -81,6 +81,9 @@ export interface StreamStats {
total_tokens: number;
input_tokens: number;
output_tokens: number;
// Breakdown of input_tokens
cached: number;
input: number;
duration_ms: number;
tool_calls: number;
}
packages/core/src/telemetry/uiTelemetry.test.ts
+4
View File
@@ -173,6 +173,7 @@ describe('UiTelemetryService', () => {
totalLatencyMs: 500,
},
tokens: {
input: 5,
prompt: 10,
candidates: 20,
total: 30,
@@ -227,6 +228,7 @@ describe('UiTelemetryService', () => {
totalLatencyMs: 1100,
},
tokens: {
input: 10,
prompt: 25,
candidates: 45,
total: 70,
@@ -301,6 +303,7 @@ describe('UiTelemetryService', () => {
totalLatencyMs: 300,
},
tokens: {
input: 0,
prompt: 0,
candidates: 0,
total: 0,
@@ -345,6 +348,7 @@ describe('UiTelemetryService', () => {
totalLatencyMs: 800,
},
tokens: {
input: 5,
prompt: 10,
candidates: 20,
total: 30,
packages/core/src/telemetry/uiTelemetry.ts
+6
View File
@@ -43,6 +43,7 @@ export interface ModelMetrics {
totalLatencyMs: number;
};
tokens: {
input: number;
prompt: number;
candidates: number;
total: number;
@@ -80,6 +81,7 @@ const createInitialModelMetrics = (): ModelMetrics => ({
totalLatencyMs: 0,
},
tokens: {
input: 0,
prompt: 0,
candidates: 0,
total: 0,
@@ -171,6 +173,10 @@ export class UiTelemetryService extends EventEmitter {
modelMetrics.tokens.cached += event.usage.cached_content_token_count;
modelMetrics.tokens.thoughts += event.usage.thoughts_token_count;
modelMetrics.tokens.tool += event.usage.tool_token_count;
modelMetrics.tokens.input = Math.max(
0,
modelMetrics.tokens.prompt - modelMetrics.tokens.cached,
);
}
private processApiError(event: ApiErrorEvent) {