repospend 0.1.6 → 0.1.8
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +24 -0
- package/README.md +11 -1
- package/dist/cli.js +507 -118
- package/docs/llms.txt +6 -1
- package/docs/pricing.md +40 -6
- package/docs/token-accounting.md +19 -2
- package/package.json +25 -7
- package/web-dist/assets/{index-B2mGGjf_.js → index-BWVpHJSG.js} +87 -87
- package/web-dist/assets/index-C3XBInjD.css +1 -0
- package/web-dist/index.html +2 -2
- package/web-dist/assets/index-Bzdfbi5w.css +0 -1
package/dist/cli.js
CHANGED
|
@@ -12,6 +12,8 @@ function summarize(sessions) {
|
|
|
12
12
|
const cacheCreationInputTokens = optionalSum(sessions, "cacheCreationInputTokens");
|
|
13
13
|
const cacheCreationInputTokens5m = optionalSum(sessions, "cacheCreationInputTokens5m");
|
|
14
14
|
const cacheCreationInputTokens1h = optionalSum(sessions, "cacheCreationInputTokens1h");
|
|
15
|
+
const cacheCreationInputTokens30m = optionalSum(sessions, "cacheCreationInputTokens30m");
|
|
16
|
+
const cacheCreationInputTokens12h = optionalSum(sessions, "cacheCreationInputTokens12h");
|
|
15
17
|
return {
|
|
16
18
|
estimatedCostUsd: cost.knownCount > 0 ? cost.value : void 0,
|
|
17
19
|
knownCostSessions: cost.knownCount,
|
|
@@ -20,13 +22,15 @@ function summarize(sessions) {
|
|
|
20
22
|
cacheCreationInputTokens,
|
|
21
23
|
cacheCreationInputTokens5m,
|
|
22
24
|
cacheCreationInputTokens1h,
|
|
25
|
+
cacheCreationInputTokens30m,
|
|
26
|
+
cacheCreationInputTokens12h,
|
|
23
27
|
inputTokens,
|
|
24
28
|
outputTokens: sum(sessions, "outputTokens"),
|
|
25
29
|
reasoningTokens: sum(sessions, "reasoningTokens"),
|
|
26
30
|
cachedInputPct: inputTokens > 0 ? cachedInputTokens / inputTokens : 0,
|
|
27
31
|
sessionCount: sessions.length,
|
|
28
32
|
repoCount: repos.length,
|
|
29
|
-
modelCount: new Set(sessions.
|
|
33
|
+
modelCount: new Set(sessions.flatMap((session) => sessionModelSlices(session).map((slice) => slice.model)).filter(Boolean)).size,
|
|
30
34
|
mostExpensiveRepo: repos.filter((repo) => repo.estimatedCostUsd !== void 0).sort((a, b) => (b.estimatedCostUsd ?? 0) - (a.estimatedCostUsd ?? 0))[0],
|
|
31
35
|
unknownCostSessions: sessions.filter((session) => session.estimatedCostUsd === void 0).length,
|
|
32
36
|
userPromptCount: optionalSum(sessions, "userPromptCount"),
|
|
@@ -57,12 +61,53 @@ function groupByRepo(sessions) {
|
|
|
57
61
|
}));
|
|
58
62
|
}
|
|
59
63
|
function groupByModel(sessions) {
|
|
60
|
-
const groups =
|
|
61
|
-
const
|
|
64
|
+
const groups = /* @__PURE__ */ new Map();
|
|
65
|
+
for (const session of sessions) {
|
|
66
|
+
for (const slice of sessionModelSlices(session)) {
|
|
67
|
+
const id = slice.model ?? "unknown-model";
|
|
68
|
+
const group = groups.get(id) ?? emptyGroup(id, slice.model ?? "Model not recorded");
|
|
69
|
+
accumulateGroup(group, session, slice);
|
|
70
|
+
groups.set(id, group);
|
|
71
|
+
}
|
|
72
|
+
}
|
|
73
|
+
const grouped = sortGroups([...groups.values()]);
|
|
74
|
+
const unknownModelSessions = sessions.filter((session) => !session.model && !(session.modelUsage ?? []).some((entry) => entry.model));
|
|
62
75
|
if (unknownModelSessions.length > 0 && unknownModelSessions.every(isClaudeSyntheticZeroUsageSession)) {
|
|
63
|
-
return
|
|
76
|
+
return grouped.map((group) => group.id === "unknown-model" ? { ...group, label: "Claude synthetic / no usage" } : group);
|
|
64
77
|
}
|
|
65
|
-
return
|
|
78
|
+
return grouped;
|
|
79
|
+
}
|
|
80
|
+
function sessionModelSlices(session) {
|
|
81
|
+
if (session.modelUsage?.length) {
|
|
82
|
+
return session.modelUsage.map((entry) => ({
|
|
83
|
+
model: entry.model,
|
|
84
|
+
inputTokens: entry.inputTokens,
|
|
85
|
+
cachedInputTokens: entry.cachedInputTokens,
|
|
86
|
+
cacheCreationInputTokens: entry.cacheCreationInputTokens,
|
|
87
|
+
cacheCreationInputTokens5m: 0,
|
|
88
|
+
cacheCreationInputTokens1h: 0,
|
|
89
|
+
cacheCreationInputTokens30m: entry.cacheCreationInputTokens30m ?? 0,
|
|
90
|
+
cacheCreationInputTokens12h: entry.cacheCreationInputTokens12h ?? 0,
|
|
91
|
+
outputTokens: entry.outputTokens,
|
|
92
|
+
reasoningTokens: entry.reasoningTokens,
|
|
93
|
+
totalTokens: entry.totalTokens,
|
|
94
|
+
estimatedCostUsd: entry.estimatedCostUsd
|
|
95
|
+
}));
|
|
96
|
+
}
|
|
97
|
+
return [{
|
|
98
|
+
model: session.model,
|
|
99
|
+
inputTokens: session.inputTokens,
|
|
100
|
+
cachedInputTokens: session.cachedInputTokens,
|
|
101
|
+
cacheCreationInputTokens: session.cacheCreationInputTokens ?? 0,
|
|
102
|
+
cacheCreationInputTokens5m: session.cacheCreationInputTokens5m ?? 0,
|
|
103
|
+
cacheCreationInputTokens1h: session.cacheCreationInputTokens1h ?? 0,
|
|
104
|
+
cacheCreationInputTokens30m: session.cacheCreationInputTokens30m ?? 0,
|
|
105
|
+
cacheCreationInputTokens12h: session.cacheCreationInputTokens12h ?? 0,
|
|
106
|
+
outputTokens: session.outputTokens,
|
|
107
|
+
reasoningTokens: session.reasoningTokens,
|
|
108
|
+
totalTokens: session.totalTokens,
|
|
109
|
+
estimatedCostUsd: session.estimatedCostUsd
|
|
110
|
+
}];
|
|
66
111
|
}
|
|
67
112
|
function groupBySourceApp(sessions) {
|
|
68
113
|
return groupBy(sessions, (session) => session.sourceApp, (session) => session.sourceApp);
|
|
@@ -77,45 +122,57 @@ function groupBy(sessions, keyFn, labelFn) {
|
|
|
77
122
|
const groups = /* @__PURE__ */ new Map();
|
|
78
123
|
for (const session of sessions) {
|
|
79
124
|
const id = keyFn(session);
|
|
80
|
-
const
|
|
81
|
-
|
|
82
|
-
id,
|
|
83
|
-
label: labelFn(session),
|
|
84
|
-
estimatedCostUsd: void 0,
|
|
85
|
-
knownCostSessions: 0,
|
|
86
|
-
unknownCostSessions: 0,
|
|
87
|
-
inputTokens: 0,
|
|
88
|
-
cachedInputTokens: 0,
|
|
89
|
-
cacheCreationInputTokens: 0,
|
|
90
|
-
cacheCreationInputTokens5m: 0,
|
|
91
|
-
cacheCreationInputTokens1h: 0,
|
|
92
|
-
outputTokens: 0,
|
|
93
|
-
reasoningTokens: 0,
|
|
94
|
-
totalTokens: 0,
|
|
95
|
-
sessionCount: 0,
|
|
96
|
-
messageCount: 0,
|
|
97
|
-
warnings: []
|
|
98
|
-
};
|
|
99
|
-
group.inputTokens += session.inputTokens;
|
|
100
|
-
group.cachedInputTokens += session.cachedInputTokens;
|
|
101
|
-
group.cacheCreationInputTokens = (group.cacheCreationInputTokens ?? 0) + (session.cacheCreationInputTokens ?? 0);
|
|
102
|
-
group.cacheCreationInputTokens5m = (group.cacheCreationInputTokens5m ?? 0) + (session.cacheCreationInputTokens5m ?? 0);
|
|
103
|
-
group.cacheCreationInputTokens1h = (group.cacheCreationInputTokens1h ?? 0) + (session.cacheCreationInputTokens1h ?? 0);
|
|
104
|
-
group.outputTokens += session.outputTokens;
|
|
105
|
-
group.reasoningTokens += session.reasoningTokens;
|
|
106
|
-
group.totalTokens += session.totalTokens;
|
|
107
|
-
group.sessionCount += 1;
|
|
108
|
-
group.messageCount += session.messageCount;
|
|
109
|
-
group.warnings = unique([...group.warnings, ...session.warnings]);
|
|
110
|
-
if (session.estimatedCostUsd !== void 0) {
|
|
111
|
-
group.estimatedCostUsd = roundCurrency((group.estimatedCostUsd ?? 0) + session.estimatedCostUsd);
|
|
112
|
-
group.knownCostSessions += 1;
|
|
113
|
-
} else {
|
|
114
|
-
group.unknownCostSessions += 1;
|
|
115
|
-
}
|
|
125
|
+
const group = groups.get(id) ?? emptyGroup(id, labelFn(session));
|
|
126
|
+
accumulateGroup(group, session, session);
|
|
116
127
|
groups.set(id, group);
|
|
117
128
|
}
|
|
118
|
-
return [...groups.values()]
|
|
129
|
+
return sortGroups([...groups.values()]);
|
|
130
|
+
}
|
|
131
|
+
function emptyGroup(id, label) {
|
|
132
|
+
return {
|
|
133
|
+
id,
|
|
134
|
+
label,
|
|
135
|
+
estimatedCostUsd: void 0,
|
|
136
|
+
knownCostSessions: 0,
|
|
137
|
+
unknownCostSessions: 0,
|
|
138
|
+
inputTokens: 0,
|
|
139
|
+
cachedInputTokens: 0,
|
|
140
|
+
cacheCreationInputTokens: 0,
|
|
141
|
+
cacheCreationInputTokens5m: 0,
|
|
142
|
+
cacheCreationInputTokens1h: 0,
|
|
143
|
+
cacheCreationInputTokens30m: 0,
|
|
144
|
+
cacheCreationInputTokens12h: 0,
|
|
145
|
+
outputTokens: 0,
|
|
146
|
+
reasoningTokens: 0,
|
|
147
|
+
totalTokens: 0,
|
|
148
|
+
sessionCount: 0,
|
|
149
|
+
messageCount: 0,
|
|
150
|
+
warnings: []
|
|
151
|
+
};
|
|
152
|
+
}
|
|
153
|
+
function accumulateGroup(group, session, slice) {
|
|
154
|
+
group.inputTokens += slice.inputTokens;
|
|
155
|
+
group.cachedInputTokens += slice.cachedInputTokens;
|
|
156
|
+
group.cacheCreationInputTokens = (group.cacheCreationInputTokens ?? 0) + (slice.cacheCreationInputTokens ?? 0);
|
|
157
|
+
group.cacheCreationInputTokens5m = (group.cacheCreationInputTokens5m ?? 0) + (slice.cacheCreationInputTokens5m ?? 0);
|
|
158
|
+
group.cacheCreationInputTokens1h = (group.cacheCreationInputTokens1h ?? 0) + (slice.cacheCreationInputTokens1h ?? 0);
|
|
159
|
+
group.cacheCreationInputTokens30m = (group.cacheCreationInputTokens30m ?? 0) + (slice.cacheCreationInputTokens30m ?? 0);
|
|
160
|
+
group.cacheCreationInputTokens12h = (group.cacheCreationInputTokens12h ?? 0) + (slice.cacheCreationInputTokens12h ?? 0);
|
|
161
|
+
group.outputTokens += slice.outputTokens;
|
|
162
|
+
group.reasoningTokens += slice.reasoningTokens;
|
|
163
|
+
group.totalTokens += slice.totalTokens;
|
|
164
|
+
group.sessionCount += 1;
|
|
165
|
+
group.messageCount += session.messageCount;
|
|
166
|
+
group.warnings = unique([...group.warnings, ...session.warnings]);
|
|
167
|
+
if (slice.estimatedCostUsd !== void 0) {
|
|
168
|
+
group.estimatedCostUsd = roundCurrency((group.estimatedCostUsd ?? 0) + slice.estimatedCostUsd);
|
|
169
|
+
group.knownCostSessions += 1;
|
|
170
|
+
} else {
|
|
171
|
+
group.unknownCostSessions += 1;
|
|
172
|
+
}
|
|
173
|
+
}
|
|
174
|
+
function sortGroups(groups) {
|
|
175
|
+
return groups.sort((a, b) => (b.estimatedCostUsd ?? 0) - (a.estimatedCostUsd ?? 0) || b.totalTokens - a.totalTokens);
|
|
119
176
|
}
|
|
120
177
|
function isClaudeSyntheticZeroUsageSession(session) {
|
|
121
178
|
return session.warnings.includes("claude_synthetic_zero_usage");
|
|
@@ -142,6 +199,8 @@ function toCsv(sessions) {
|
|
|
142
199
|
"cacheCreationInputTokens",
|
|
143
200
|
"cacheCreationInputTokens5m",
|
|
144
201
|
"cacheCreationInputTokens1h",
|
|
202
|
+
"cacheCreationInputTokens30m",
|
|
203
|
+
"cacheCreationInputTokens12h",
|
|
145
204
|
"outputTokens",
|
|
146
205
|
"reasoningTokens",
|
|
147
206
|
"reasoningOutputTokens",
|
|
@@ -253,9 +312,11 @@ function claudePricingFamilyModel(model, isUsable = () => true) {
|
|
|
253
312
|
"claude-sonnet-4-5",
|
|
254
313
|
"claude-sonnet-4",
|
|
255
314
|
"claude-haiku-4-5",
|
|
315
|
+
"claude-haiku-5-5",
|
|
256
316
|
"claude-3-5-haiku"
|
|
257
317
|
];
|
|
258
318
|
const normalized = normalizePricingModelId(model);
|
|
319
|
+
if (normalized.includes("-long-context")) return void 0;
|
|
259
320
|
const matchesFamily = (candidate) => normalized === candidate || normalized.startsWith(`${candidate}-`) || normalized.startsWith(`${candidate}.`);
|
|
260
321
|
if (normalized.includes("fast-mode")) {
|
|
261
322
|
return families.find((candidate) => candidate.includes("fast-mode") && matchesFamily(candidate) && isUsable(candidate));
|
|
@@ -271,7 +332,7 @@ function copilotPricingFamilyModel(model, isUsable = () => true) {
|
|
|
271
332
|
if ((normalized === "gemini-3-flash" || normalized.startsWith("gemini-3-flash-")) && isUsable("gemini-3-flash")) return "gemini-3-flash";
|
|
272
333
|
if ((normalized === "gemini-3.1-pro" || normalized.startsWith("gemini-3.1-pro-")) && isUsable("gemini-3.1-pro")) return "gemini-3.1-pro";
|
|
273
334
|
if ((normalized === "gemini-3.5-flash" || normalized.startsWith("gemini-3.5-flash-")) && isUsable("gemini-3.5-flash")) return "gemini-3.5-flash";
|
|
274
|
-
for (const candidate of ["gemini-3.6-flash", "gemini-3.7-flash", "gemini-3.8-flash", "grok-4.5", "grok-4.6", "grok-4.7", "mai-code-1.1-flash", "kimi-k3"]) {
|
|
335
|
+
for (const candidate of ["gemini-3.6-flash", "gemini-3.7-flash", "gemini-3.8-flash", "gemini-4-argon", "grok-4.5", "grok-4.6", "grok-4.7", "mai-code-1.1-flash", "kimi-k3"]) {
|
|
275
336
|
if ((normalized === candidate || normalized.startsWith(`${candidate}-`)) && isUsable(candidate)) return candidate;
|
|
276
337
|
}
|
|
277
338
|
if ((normalized === "raptor-mini" || normalized.startsWith("raptor-mini-") || normalized.startsWith("oswe-vscode")) && isUsable("raptor-mini")) return "raptor-mini";
|
|
@@ -282,7 +343,7 @@ function copilotPricingFamilyModel(model, isUsable = () => true) {
|
|
|
282
343
|
return void 0;
|
|
283
344
|
}
|
|
284
345
|
function parseVersionedModel(normalized) {
|
|
285
|
-
if (normalized.includes("fast-mode")) return void 0;
|
|
346
|
+
if (normalized.includes("fast-mode") || normalized.includes("-long-context")) return void 0;
|
|
286
347
|
const claude = normalized.match(/^(claude-(?:fable|mythos|opus|sonnet|haiku))-(\d+)(?:-(\d{1,2}))?(?!\d)/);
|
|
287
348
|
if (claude) {
|
|
288
349
|
const minor = claude[3] ? Number(claude[3]) : 0;
|
|
@@ -318,17 +379,36 @@ function positiveRate(value) {
|
|
|
318
379
|
function isUsableModelPricing(pricing) {
|
|
319
380
|
return positiveRate(pricing?.inputPerMillion) && positiveRate(pricing?.outputPerMillion);
|
|
320
381
|
}
|
|
321
|
-
function
|
|
382
|
+
function resolvePricingForModel(model, table) {
|
|
322
383
|
const lower = normalizePricingModelId(model);
|
|
323
|
-
const
|
|
324
|
-
|
|
325
|
-
|
|
326
|
-
|
|
327
|
-
|
|
328
|
-
|
|
329
|
-
|
|
330
|
-
|
|
331
|
-
|
|
384
|
+
const exactModel = table[model] ? model : table[lower] ? lower : void 0;
|
|
385
|
+
const exactPricing = exactModel ? table[exactModel] : void 0;
|
|
386
|
+
if (isUsableModelPricing(exactPricing)) {
|
|
387
|
+
return { pricing: exactPricing, sourceModel: exactModel, inherited: false };
|
|
388
|
+
}
|
|
389
|
+
const versioned = versionedFallbackModel(lower, Object.keys(table), (candidate) => isUsableModelPricing(table[candidate]));
|
|
390
|
+
if (versioned) {
|
|
391
|
+
return { pricing: table[versioned], sourceModel: versioned, inherited: versioned !== model && versioned !== lower };
|
|
392
|
+
}
|
|
393
|
+
const claudeFamily = claudePricingFamilyModel(lower, (candidate) => isUsableModelPricing(table[candidate]));
|
|
394
|
+
if (claudeFamily) {
|
|
395
|
+
return { pricing: table[claudeFamily], sourceModel: claudeFamily, inherited: claudeFamily !== model && claudeFamily !== lower };
|
|
396
|
+
}
|
|
397
|
+
const copilotFamily = copilotPricingFamilyModel(lower, (candidate) => isUsableModelPricing(table[candidate]));
|
|
398
|
+
if (copilotFamily) {
|
|
399
|
+
return { pricing: table[copilotFamily], sourceModel: copilotFamily, inherited: copilotFamily !== model && copilotFamily !== lower };
|
|
400
|
+
}
|
|
401
|
+
return {
|
|
402
|
+
pricing: exactPricing ?? {
|
|
403
|
+
inputPerMillion: 0,
|
|
404
|
+
cacheCreationInputPerMillion: 0,
|
|
405
|
+
cachedInputPerMillion: 0,
|
|
406
|
+
outputPerMillion: 0,
|
|
407
|
+
reasoningOutputPerMillion: 0
|
|
408
|
+
},
|
|
409
|
+
sourceModel: exactModel,
|
|
410
|
+
inherited: false
|
|
411
|
+
};
|
|
332
412
|
}
|
|
333
413
|
|
|
334
414
|
// packages/core/src/pricing.ts
|
|
@@ -339,30 +419,47 @@ var pricingInfo = {
|
|
|
339
419
|
{ label: "OpenAI pricing reference", url: "https://developers.openai.com/api/docs/pricing" },
|
|
340
420
|
{ label: "OpenAI GPT-6 model reference", url: "https://developers.openai.com/api/docs/models" },
|
|
341
421
|
{ label: "OpenAI GPT-6.1 Sol reference", url: "https://developers.openai.com/api/docs/models/gpt-6.1-sol" },
|
|
422
|
+
{ label: "OpenAI GPT-6.1 Terra reference", url: "https://developers.openai.com/api/docs/models/gpt-6.1-terra" },
|
|
423
|
+
{ label: "OpenAI prompt caching guide", url: "https://developers.openai.com/api/docs/guides/prompt-caching" },
|
|
342
424
|
{ label: "OpenAI GPT-5.6 preview pricing", url: "https://openai.com/index/previewing-gpt-5-6-sol/" },
|
|
343
425
|
{ label: "OpenAI GPT-5.6 price update", url: "https://openai.com/index/advancing-the-price-performance-frontier-with-gpt-5-6/" },
|
|
344
426
|
{ label: "Claude pricing reference", url: "https://platform.claude.com/docs/en/about-claude/pricing" },
|
|
427
|
+
{ label: "Claude Haiku 5.5 and Sonnet cache update", url: "https://www.anthropic.com/claude-haiku-5-5" },
|
|
428
|
+
{ label: "Gemini 4 Argon pricing", url: "https://ai.google.dev/gemini-api/docs/latest-model?hl=en" },
|
|
345
429
|
{ label: "GitHub Copilot model pricing reference", url: "https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing" }
|
|
346
430
|
],
|
|
347
431
|
unit: "USD per 1M tokens",
|
|
348
|
-
updatedAt: "2026-
|
|
349
|
-
note: "RepoSpend estimates API-equivalent cost from local token counts and public Standard pricing.
|
|
432
|
+
updatedAt: "2026-10-27",
|
|
433
|
+
note: "RepoSpend estimates API-equivalent cost from local token counts and public Standard pricing. Requests over each model's long-context threshold reprice per request when request sizes are recorded; aggregate-only usage assumes short-context rates and is flagged. Claude transcript fast-mode speeds use explicit published cards. Other processing modes, regional, and account-specific pricing can differ. This is not your actual bill."
|
|
350
434
|
};
|
|
351
435
|
var defaultPricing = {
|
|
352
|
-
"gpt-6-astra": { inputPerMillion: 10, cachedInputPerMillion: 1, cacheCreationInputPerMillion: 12.5, outputPerMillion: 50, reasoningOutputPerMillion: 50 },
|
|
353
|
-
"gpt-6-
|
|
354
|
-
"gpt-6
|
|
355
|
-
"gpt-6-
|
|
356
|
-
"gpt-
|
|
357
|
-
"gpt-
|
|
358
|
-
"gpt-
|
|
359
|
-
"gpt-
|
|
436
|
+
"gpt-6-astra": { inputPerMillion: 10, cachedInputPerMillion: 1, cacheCreationInputPerMillion: 12.5, cacheCreationInput30mPerMillion: 12.5, cacheCreationInput12hPerMillion: 25, outputPerMillion: 50, reasoningOutputPerMillion: 50 },
|
|
437
|
+
"gpt-6-astra-long-context": { inputPerMillion: 20, cachedInputPerMillion: 2, cacheCreationInputPerMillion: 25, cacheCreationInput30mPerMillion: 25, cacheCreationInput12hPerMillion: 50, outputPerMillion: 75, reasoningOutputPerMillion: 75, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole request when a single API call exceeds 272K input tokens: input and cache rates 2x, output 1.5x." },
|
|
438
|
+
"gpt-6-sol": { inputPerMillion: 2, cachedInputPerMillion: 0.2, cacheCreationInputPerMillion: 2.5, cacheCreationInput30mPerMillion: 2.5, cacheCreationInput12hPerMillion: 5, outputPerMillion: 10, reasoningOutputPerMillion: 10 },
|
|
439
|
+
"gpt-6-sol-long-context": { inputPerMillion: 4, cachedInputPerMillion: 0.4, cacheCreationInputPerMillion: 5, cacheCreationInput30mPerMillion: 5, cacheCreationInput12hPerMillion: 10, outputPerMillion: 15, reasoningOutputPerMillion: 15, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole request when a single API call exceeds 272K input tokens: input and cache rates 2x, output 1.5x." },
|
|
440
|
+
"gpt-6-luna": { inputPerMillion: 0.1, cachedInputPerMillion: 0.01, cacheCreationInputPerMillion: 0.125, cacheCreationInput30mPerMillion: 0.125, cacheCreationInput12hPerMillion: 0.25, outputPerMillion: 0.5, reasoningOutputPerMillion: 0.5 },
|
|
441
|
+
"gpt-6-luna-long-context": { inputPerMillion: 0.2, cachedInputPerMillion: 0.02, cacheCreationInputPerMillion: 0.25, cacheCreationInput30mPerMillion: 0.25, cacheCreationInput12hPerMillion: 0.5, outputPerMillion: 0.75, reasoningOutputPerMillion: 0.75, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole request when a single API call exceeds 272K input tokens: input and cache rates 2x, output 1.5x." },
|
|
442
|
+
"gpt-6.1-sol": { inputPerMillion: 2, cachedInputPerMillion: 0.1, cacheCreationInputPerMillion: 2.5, cacheCreationInput30mPerMillion: 2.5, cacheCreationInput12hPerMillion: 5, outputPerMillion: 10, reasoningOutputPerMillion: 10 },
|
|
443
|
+
"gpt-6.1-sol-long-context": { inputPerMillion: 4, cachedInputPerMillion: 0.2, cacheCreationInputPerMillion: 5, cacheCreationInput30mPerMillion: 5, cacheCreationInput12hPerMillion: 10, outputPerMillion: 15, reasoningOutputPerMillion: 15, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole request when a single API call exceeds 272K input tokens: input and cache rates 2x, output 1.5x." },
|
|
444
|
+
"gpt-6.1-terra": { inputPerMillion: 1, cachedInputPerMillion: 0.05, cacheCreationInputPerMillion: 1.25, cacheCreationInput30mPerMillion: 1.25, cacheCreationInput12hPerMillion: 2.5, outputPerMillion: 5, reasoningOutputPerMillion: 5, note: "GPT-6.1 Terra Standard short-context rates, released October 26, 2026. Cache writes are 1.25x input for 30-minute TTL and 2.5x for 12-hour TTL." },
|
|
445
|
+
"gpt-6.1-terra-long-context": { inputPerMillion: 2, cachedInputPerMillion: 0.1, cacheCreationInputPerMillion: 2.5, cacheCreationInput30mPerMillion: 2.5, cacheCreationInput12hPerMillion: 5, outputPerMillion: 7.5, reasoningOutputPerMillion: 7.5, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole request when a single API call exceeds 272K input tokens: input and cache rates 2x, output 1.5x." },
|
|
446
|
+
"gpt-5.6": { inputPerMillion: 4, cachedInputPerMillion: 0.4, cacheCreationInputPerMillion: 5, cacheCreationInput30mPerMillion: 5, cacheCreationInput12hPerMillion: 10, outputPerMillion: 20, reasoningOutputPerMillion: 20, note: "GPT-5.6 aliases Sol. Promotional Standard short-context rates are available at least through November 21, 2026: $4 input / $20 output, $0.40 cache reads, and $5 cache writes per 1M tokens." },
|
|
447
|
+
"gpt-5.6-sol": { inputPerMillion: 4, cachedInputPerMillion: 0.4, cacheCreationInputPerMillion: 5, cacheCreationInput30mPerMillion: 5, cacheCreationInput12hPerMillion: 10, outputPerMillion: 20, reasoningOutputPerMillion: 20, note: "Promotional Standard short-context rates available at least through November 21, 2026." },
|
|
448
|
+
"gpt-5.6-terra": { inputPerMillion: 2, cachedInputPerMillion: 0.2, cacheCreationInputPerMillion: 2.5, cacheCreationInput30mPerMillion: 2.5, cacheCreationInput12hPerMillion: 5, outputPerMillion: 12, reasoningOutputPerMillion: 12, note: "GPT-5.6 Terra reduced API pricing effective July 30, 2026. Cache reads are 90% below input and cache writes are 1.25x input." },
|
|
449
|
+
"gpt-5.6-luna": { inputPerMillion: 0.2, cachedInputPerMillion: 0.02, cacheCreationInputPerMillion: 0.25, cacheCreationInput30mPerMillion: 0.25, cacheCreationInput12hPerMillion: 0.5, outputPerMillion: 1.2, reasoningOutputPerMillion: 1.2, note: "GPT-5.6 Luna reduced API pricing effective July 30, 2026. Cache reads are 90% below input and cache writes are 1.25x input." },
|
|
360
450
|
"gpt-5.5": { inputPerMillion: 5, cachedInputPerMillion: 0.5, outputPerMillion: 30, reasoningOutputPerMillion: 30 },
|
|
361
451
|
"gpt-5.5-pro": { inputPerMillion: 30, outputPerMillion: 180, reasoningOutputPerMillion: 180 },
|
|
362
452
|
"gpt-5.4": { inputPerMillion: 2.5, cachedInputPerMillion: 0.25, outputPerMillion: 15, reasoningOutputPerMillion: 15 },
|
|
363
453
|
"gpt-5.4-mini": { inputPerMillion: 0.75, cachedInputPerMillion: 0.075, outputPerMillion: 4.5, reasoningOutputPerMillion: 4.5 },
|
|
364
454
|
"gpt-5.4-nano": { inputPerMillion: 0.2, cachedInputPerMillion: 0.02, outputPerMillion: 1.25, reasoningOutputPerMillion: 1.25 },
|
|
365
455
|
"gpt-5.4-pro": { inputPerMillion: 30, outputPerMillion: 180, reasoningOutputPerMillion: 180 },
|
|
456
|
+
"gpt-5.6-sol-long-context": { inputPerMillion: 8, cachedInputPerMillion: 0.8, cacheCreationInputPerMillion: 10, cacheCreationInput30mPerMillion: 10, cacheCreationInput12hPerMillion: 20, outputPerMillion: 30, reasoningOutputPerMillion: 30, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole GPT-5.6 request when a single API call exceeds 272K input tokens: input and cache rates 2x, output 1.5x." },
|
|
457
|
+
"gpt-5.6-terra-long-context": { inputPerMillion: 4, cachedInputPerMillion: 0.4, cacheCreationInputPerMillion: 5, cacheCreationInput30mPerMillion: 5, cacheCreationInput12hPerMillion: 10, outputPerMillion: 18, reasoningOutputPerMillion: 18, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole GPT-5.6 Terra request when a single API call exceeds 272K input tokens: input and cache rates 2x, output 1.5x." },
|
|
458
|
+
"gpt-5.6-luna-long-context": { inputPerMillion: 0.4, cachedInputPerMillion: 0.04, cacheCreationInputPerMillion: 0.5, cacheCreationInput30mPerMillion: 0.5, cacheCreationInput12hPerMillion: 1, outputPerMillion: 1.8, reasoningOutputPerMillion: 1.8, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole GPT-5.6 Luna request when a single API call exceeds 272K input tokens: input and cache rates 2x, output 1.5x." },
|
|
459
|
+
"gpt-5.5-long-context": { inputPerMillion: 10, cachedInputPerMillion: 1, outputPerMillion: 45, reasoningOutputPerMillion: 45, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole GPT-5.5 request when a single API call exceeds 272K input tokens. No published cache-write rates." },
|
|
460
|
+
"gpt-5.5-pro-long-context": { inputPerMillion: 60, outputPerMillion: 270, reasoningOutputPerMillion: 270, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole GPT-5.5 Pro request when a single API call exceeds 272K input tokens. No published cache rates." },
|
|
461
|
+
"gpt-5.4-long-context": { inputPerMillion: 5, cachedInputPerMillion: 0.5, outputPerMillion: 22.5, reasoningOutputPerMillion: 22.5, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole GPT-5.4 request when a single API call exceeds 272K input tokens. No published cache-write rates." },
|
|
462
|
+
"gpt-5.4-pro-long-context": { inputPerMillion: 60, outputPerMillion: 270, reasoningOutputPerMillion: 270, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole GPT-5.4 Pro request when a single API call exceeds 272K input tokens. No published cache rates." },
|
|
366
463
|
"gpt-5.3-codex": { inputPerMillion: 1.75, cachedInputPerMillion: 0.175, outputPerMillion: 14, reasoningOutputPerMillion: 14 },
|
|
367
464
|
"gpt-5.3-codex-spark": { inputPerMillion: 1.75, cachedInputPerMillion: 0.175, outputPerMillion: 14, reasoningOutputPerMillion: 14, note: "Research preview pricing is not final in the OpenAI Codex rate card; this API-equivalent rate matches the practical ccusage/LiteLLM-style estimate." },
|
|
368
465
|
"gpt-5.2": { inputPerMillion: 1.75, cachedInputPerMillion: 0.175, outputPerMillion: 14, reasoningOutputPerMillion: 14 },
|
|
@@ -387,6 +484,7 @@ var defaultPricing = {
|
|
|
387
484
|
"gemini-3.6-flash": { inputPerMillion: 0.75, cachedInputPerMillion: 0.075, outputPerMillion: 3.75, note: "GitHub Copilot promotional rates through December 31, 2026; API-equivalent estimate." },
|
|
388
485
|
"gemini-3.7-flash": { inputPerMillion: 0.75, cachedInputPerMillion: 0.075, outputPerMillion: 3.75, note: "GitHub Copilot promotional rates through December 31, 2026; API-equivalent estimate." },
|
|
389
486
|
"gemini-3.8-flash": { inputPerMillion: 0.75, cachedInputPerMillion: 0.075, outputPerMillion: 3.75, note: "GitHub Copilot promotional rates through December 31, 2026; API-equivalent estimate." },
|
|
487
|
+
"gemini-4-argon": { inputPerMillion: 2, cachedInputPerMillion: 0.1, outputPerMillion: 10, reasoningOutputPerMillion: 10, note: "Introductory rates through November 30, 2026. From December 1: $4 input / $0.20 cached input / $20 output (including thinking) per 1M tokens. No explicit cache-write charge. Bundled prices are a dated snapshot, not an automatic future rate switch." },
|
|
390
488
|
"grok-4.5": { inputPerMillion: 2, cachedInputPerMillion: 0.5, outputPerMillion: 6, note: "GitHub Copilot Standard rates up to 200K input tokens; longer requests use different rates." },
|
|
391
489
|
"grok-4.6": { inputPerMillion: 2, cachedInputPerMillion: 0.5, outputPerMillion: 6, note: "GitHub Copilot Standard rates up to 200K input tokens; longer requests use different rates." },
|
|
392
490
|
"grok-4.7": { inputPerMillion: 2, cachedInputPerMillion: 0.5, outputPerMillion: 6, note: "GitHub Copilot Standard rates up to 200K input tokens; longer requests use different rates." },
|
|
@@ -414,15 +512,18 @@ var defaultPricing = {
|
|
|
414
512
|
"claude-opus-4-5": { inputPerMillion: 5, cacheCreationInput5mPerMillion: 6.25, cacheCreationInput1hPerMillion: 10, cacheCreationInputPerMillion: 10, cachedInputPerMillion: 0.5, outputPerMillion: 25 },
|
|
415
513
|
"claude-opus-4-1": { inputPerMillion: 15, cacheCreationInput5mPerMillion: 18.75, cacheCreationInput1hPerMillion: 30, cacheCreationInputPerMillion: 30, cachedInputPerMillion: 1.5, outputPerMillion: 75 },
|
|
416
514
|
"claude-opus-4": { inputPerMillion: 15, cacheCreationInput5mPerMillion: 18.75, cacheCreationInput1hPerMillion: 30, cacheCreationInputPerMillion: 30, cachedInputPerMillion: 1.5, outputPerMillion: 75 },
|
|
417
|
-
"claude-sonnet-5-5": { inputPerMillion: 2, cacheCreationInput5mPerMillion: 2.5, cacheCreationInput1hPerMillion: 4, cacheCreationInputPerMillion: 4, cachedInputPerMillion: 0.
|
|
515
|
+
"claude-sonnet-5-5": { inputPerMillion: 2, cacheCreationInput5mPerMillion: 2.5, cacheCreationInput1hPerMillion: 4, cacheCreationInputPerMillion: 4, cachedInputPerMillion: 0.1, outputPerMillion: 10 },
|
|
418
516
|
"claude-sonnet-5": { inputPerMillion: 2, cacheCreationInput5mPerMillion: 2.5, cacheCreationInput1hPerMillion: 4, cacheCreationInputPerMillion: 4, cachedInputPerMillion: 0.2, outputPerMillion: 10, note: "Anthropic made the $2 input / $10 output introductory pricing permanent on August 10, 2026." },
|
|
419
517
|
"claude-sonnet-4-6": { inputPerMillion: 3, cacheCreationInput5mPerMillion: 3.75, cacheCreationInput1hPerMillion: 6, cacheCreationInputPerMillion: 6, cachedInputPerMillion: 0.3, outputPerMillion: 15 },
|
|
420
518
|
"claude-sonnet-4-5": { inputPerMillion: 3, cacheCreationInput5mPerMillion: 3.75, cacheCreationInput1hPerMillion: 6, cacheCreationInputPerMillion: 6, cachedInputPerMillion: 0.3, outputPerMillion: 15 },
|
|
421
519
|
"claude-sonnet-4": { inputPerMillion: 3, cacheCreationInput5mPerMillion: 3.75, cacheCreationInput1hPerMillion: 6, cacheCreationInputPerMillion: 6, cachedInputPerMillion: 0.3, outputPerMillion: 15 },
|
|
422
520
|
"claude-haiku-4-5": { inputPerMillion: 1, cacheCreationInput5mPerMillion: 1.25, cacheCreationInput1hPerMillion: 2, cacheCreationInputPerMillion: 2, cachedInputPerMillion: 0.1, outputPerMillion: 5 },
|
|
521
|
+
"claude-haiku-5-5": { inputPerMillion: 0.1, cacheCreationInput5mPerMillion: 0.125, cacheCreationInput1hPerMillion: 0.2, cacheCreationInputPerMillion: 0.2, cachedInputPerMillion: 0.01, outputPerMillion: 0.5, note: "Prompts up to 100,000 input tokens per request, including cache reads and writes. Larger prompts use the separately editable claude-haiku-5-5-long-context card. Aggregate-only usage stays unpriced when request lengths are unknown." },
|
|
522
|
+
"claude-haiku-5-5-long-context": { inputPerMillion: 0.5, cacheCreationInput5mPerMillion: 0.625, cacheCreationInput1hPerMillion: 1, cacheCreationInputPerMillion: 1, cachedInputPerMillion: 0.05, outputPerMillion: 2.5, note: "RepoSpend pricing card, not a provider model ID. Applied to the whole Haiku 5.5 request when total prompt length exceeds 100,000 tokens." },
|
|
423
523
|
"claude-3-5-haiku": { inputPerMillion: 0.8, cacheCreationInput5mPerMillion: 1, cacheCreationInput1hPerMillion: 1.6, cacheCreationInputPerMillion: 1.6, cachedInputPerMillion: 0.08, outputPerMillion: 4 }
|
|
424
524
|
};
|
|
425
525
|
var legacyBundledPricing = {
|
|
526
|
+
"claude-sonnet-5-5": { inputPerMillion: 2, cacheCreationInput5mPerMillion: 2.5, cacheCreationInput1hPerMillion: 4, cacheCreationInputPerMillion: 4, cachedInputPerMillion: 0.2, outputPerMillion: 10 },
|
|
426
527
|
"gpt-5.6": { inputPerMillion: 5, cachedInputPerMillion: 0.5, cacheCreationInputPerMillion: 6.25, outputPerMillion: 30, reasoningOutputPerMillion: 30, note: "GPT-5.6 Sol flagship tier. OpenAI preview pricing lists Sol at $5 input / $30 output per 1M tokens, cache writes at 1.25x input, and cache reads at a 90% discount." },
|
|
427
528
|
"gpt-5.6-sol": { inputPerMillion: 5, cachedInputPerMillion: 0.5, cacheCreationInputPerMillion: 6.25, outputPerMillion: 30, reasoningOutputPerMillion: 30 },
|
|
428
529
|
"gpt-5.6-terra": { inputPerMillion: 2.5, cachedInputPerMillion: 0.25, cacheCreationInputPerMillion: 3.125, outputPerMillion: 15, reasoningOutputPerMillion: 15 },
|
|
@@ -529,17 +630,94 @@ function savePricingTable(pricingPath, pricing) {
|
|
|
529
630
|
function repospendHome() {
|
|
530
631
|
return process.env.REPOSPEND_HOME || path.join(os.homedir(), ".repospend");
|
|
531
632
|
}
|
|
532
|
-
function calculateCostUsd(usage, pricing) {
|
|
633
|
+
function calculateCostUsd(usage, pricing, requestInputTokens) {
|
|
634
|
+
const cost = unroundedCostUsd(usage, pricing, requestInputTokens);
|
|
635
|
+
return cost === void 0 ? void 0 : Number((cost + Number.EPSILON).toFixed(6));
|
|
636
|
+
}
|
|
637
|
+
function calculateRequestCosts(requests, pricing) {
|
|
638
|
+
const resolutions = /* @__PURE__ */ new Map();
|
|
639
|
+
const warnings = /* @__PURE__ */ new Set();
|
|
640
|
+
const assumptions = /* @__PURE__ */ new Set();
|
|
641
|
+
let total = 0;
|
|
642
|
+
const requestCosts = requests.map((request) => {
|
|
643
|
+
let resolution = request.model ? resolutions.get(request.model) : void 0;
|
|
644
|
+
if (request.model && !resolution) {
|
|
645
|
+
resolution = resolvePricingForModel(request.model, pricing);
|
|
646
|
+
resolutions.set(request.model, resolution);
|
|
647
|
+
}
|
|
648
|
+
const context = requestContextPricing(request, pricing, resolution);
|
|
649
|
+
if (context.assumption) assumptions.add(context.assumption);
|
|
650
|
+
const cost = context.pricing === void 0 ? void 0 : costFromModelPricing(request, context.pricing);
|
|
651
|
+
if (cost !== void 0) total += cost;
|
|
652
|
+
else if (request.model && resolution) {
|
|
653
|
+
warnings.add(context.unknownContext ? "unknown_request_context" : `unknown_pricing_model:${normalizePricingModelId(request.model)}`);
|
|
654
|
+
} else warnings.add("unknown_pricing");
|
|
655
|
+
return cost;
|
|
656
|
+
});
|
|
657
|
+
return {
|
|
658
|
+
costUsd: requests.length && !warnings.size ? Number((total + Number.EPSILON).toFixed(6)) : void 0,
|
|
659
|
+
requestCosts,
|
|
660
|
+
warnings: [...warnings],
|
|
661
|
+
assumptions: [...assumptions]
|
|
662
|
+
};
|
|
663
|
+
}
|
|
664
|
+
var REQUEST_CONTEXT_RULES = [
|
|
665
|
+
{ baseModel: "claude-haiku-5-5", thresholdInputTokens: 1e5, longContextModel: "claude-haiku-5-5-long-context", unknownRequestSize: "unpriced", missingLongContextCard: "unpriced", inheritedModelPattern: /^claude-haiku-5-5(?:$|[-.@[])/ },
|
|
666
|
+
{ baseModel: "gpt-6-astra", thresholdInputTokens: 272e3, longContextModel: "gpt-6-astra-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
667
|
+
{ baseModel: "gpt-6-sol", thresholdInputTokens: 272e3, longContextModel: "gpt-6-sol-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
668
|
+
{ baseModel: "gpt-6-luna", thresholdInputTokens: 272e3, longContextModel: "gpt-6-luna-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
669
|
+
{ baseModel: "gpt-6.1-sol", thresholdInputTokens: 272e3, longContextModel: "gpt-6.1-sol-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
670
|
+
{ baseModel: "gpt-6.1-terra", thresholdInputTokens: 272e3, longContextModel: "gpt-6.1-terra-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
671
|
+
// GPT-5.x Standard cards carry the same >272K tiers; mini/nano have no published tier.
|
|
672
|
+
{ baseModel: "gpt-5.6", thresholdInputTokens: 272e3, longContextModel: "gpt-5.6-sol-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
673
|
+
{ baseModel: "gpt-5.6-sol", thresholdInputTokens: 272e3, longContextModel: "gpt-5.6-sol-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
674
|
+
{ baseModel: "gpt-5.6-terra", thresholdInputTokens: 272e3, longContextModel: "gpt-5.6-terra-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
675
|
+
{ baseModel: "gpt-5.6-luna", thresholdInputTokens: 272e3, longContextModel: "gpt-5.6-luna-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
676
|
+
{ baseModel: "gpt-5.5", thresholdInputTokens: 272e3, longContextModel: "gpt-5.5-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
677
|
+
{ baseModel: "gpt-5.5-pro", thresholdInputTokens: 272e3, longContextModel: "gpt-5.5-pro-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
678
|
+
{ baseModel: "gpt-5.4", thresholdInputTokens: 272e3, longContextModel: "gpt-5.4-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true },
|
|
679
|
+
{ baseModel: "gpt-5.4-pro", thresholdInputTokens: 272e3, longContextModel: "gpt-5.4-pro-long-context", unknownRequestSize: "short-context", missingLongContextCard: "base", canonicalFlatOverride: true }
|
|
680
|
+
];
|
|
681
|
+
function pricingContextRules() {
|
|
682
|
+
return REQUEST_CONTEXT_RULES.map(({ baseModel, longContextModel, thresholdInputTokens }) => ({ baseModel, longContextModel, thresholdInputTokens }));
|
|
683
|
+
}
|
|
684
|
+
function requestContextPricing(request, pricing, resolution) {
|
|
685
|
+
if (!request.model || !resolution) return { pricing: void 0, unknownContext: false };
|
|
686
|
+
const normalizedModel = normalizePricingModelId(request.model);
|
|
687
|
+
const rule = REQUEST_CONTEXT_RULES.find((candidate) => candidate.baseModel === resolution.sourceModel || resolution.inherited && candidate.inheritedModelPattern?.test(normalizedModel));
|
|
688
|
+
if (!rule) return { pricing: resolution.pricing, unknownContext: false };
|
|
689
|
+
const baseRow = pricing[rule.baseModel];
|
|
690
|
+
const basePricing = resolution.sourceModel === rule.baseModel ? resolution.pricing : isUsableModelPricing(baseRow) ? baseRow : void 0;
|
|
691
|
+
const requestInputTokens = request.requestInputTokens;
|
|
692
|
+
if (requestInputTokens === void 0 || !Number.isFinite(requestInputTokens) || requestInputTokens < 0) {
|
|
693
|
+
return rule.unknownRequestSize === "unpriced" ? { pricing: void 0, unknownContext: true } : { pricing: basePricing, unknownContext: false, assumption: `assumed_short_context_rates:${rule.baseModel}` };
|
|
694
|
+
}
|
|
695
|
+
if (requestInputTokens <= rule.thresholdInputTokens) return { pricing: basePricing, unknownContext: false };
|
|
696
|
+
const bundledBase = defaultPricing[rule.baseModel];
|
|
697
|
+
if (rule.canonicalFlatOverride && resolution.sourceModel === rule.baseModel && bundledBase && basePricing && !pricingEntriesEqual(basePricing, bundledBase) && pricingEntriesEqual(pricing[rule.longContextModel], defaultPricing[rule.longContextModel])) {
|
|
698
|
+
return { pricing: basePricing, unknownContext: false };
|
|
699
|
+
}
|
|
700
|
+
const longContext = resolvePricingForModel(rule.longContextModel, pricing);
|
|
701
|
+
if (isUsableModelPricing(longContext.pricing)) return { pricing: longContext.pricing, unknownContext: false };
|
|
702
|
+
return rule.missingLongContextCard === "unpriced" ? { pricing: void 0, unknownContext: false } : { pricing: basePricing, unknownContext: false, assumption: `missing_long_context_pricing:${rule.baseModel}` };
|
|
703
|
+
}
|
|
704
|
+
function unroundedCostUsd(usage, pricing, requestInputTokens) {
|
|
533
705
|
if (!usage.model) {
|
|
534
706
|
return void 0;
|
|
535
707
|
}
|
|
536
|
-
const
|
|
537
|
-
|
|
708
|
+
const resolution = resolvePricingForModel(usage.model, pricing);
|
|
709
|
+
const modelPricing = requestContextPricing({ ...usage, requestInputTokens }, pricing, resolution).pricing;
|
|
710
|
+
return costFromModelPricing(usage, modelPricing);
|
|
711
|
+
}
|
|
712
|
+
function costFromModelPricing(usage, modelPricing) {
|
|
713
|
+
if (!usage.model || !isUsableModelPricing(modelPricing)) {
|
|
538
714
|
return void 0;
|
|
539
715
|
}
|
|
540
716
|
const cacheCreationInput5m = usage.cacheCreationInputTokens5m ?? 0;
|
|
541
717
|
const cacheCreationInput1h = usage.cacheCreationInputTokens1h ?? 0;
|
|
542
|
-
const
|
|
718
|
+
const cacheCreationInput30m = usage.cacheCreationInputTokens30m ?? 0;
|
|
719
|
+
const cacheCreationInput12h = usage.cacheCreationInputTokens12h ?? 0;
|
|
720
|
+
const splitCacheCreationInput = cacheCreationInput5m + cacheCreationInput1h + cacheCreationInput30m + cacheCreationInput12h;
|
|
543
721
|
const cacheCreationInput = usage.cacheCreationInputTokens ?? splitCacheCreationInput;
|
|
544
722
|
const unclassifiedCacheCreationInput = Math.max(cacheCreationInput - splitCacheCreationInput, 0);
|
|
545
723
|
const billableInput = Math.max(usage.inputTokens - usage.cachedInputTokens - cacheCreationInput, 0);
|
|
@@ -547,16 +725,18 @@ function calculateCostUsd(usage, pricing) {
|
|
|
547
725
|
const cacheCreationRate = firstPositiveRate(modelPricing.cacheCreationInputPerMillion, modelPricing.cacheCreationInput1hPerMillion, modelPricing.cacheCreationInput5mPerMillion, modelPricing.inputPerMillion);
|
|
548
726
|
const cacheCreation5mRate = firstPositiveRate(modelPricing.cacheCreationInput5mPerMillion, modelPricing.cacheCreationInputPerMillion, modelPricing.inputPerMillion);
|
|
549
727
|
const cacheCreation1hRate = firstPositiveRate(modelPricing.cacheCreationInput1hPerMillion, modelPricing.cacheCreationInputPerMillion, modelPricing.inputPerMillion);
|
|
728
|
+
const cacheCreation30mRate = firstPositiveRate(modelPricing.cacheCreationInput30mPerMillion, modelPricing.cacheCreationInputPerMillion, modelPricing.inputPerMillion);
|
|
729
|
+
const cacheCreation12hRate = firstPositiveRate(modelPricing.cacheCreationInput12hPerMillion, modelPricing.cacheCreationInputPerMillion, modelPricing.inputPerMillion);
|
|
550
730
|
const reasoningRate = modelPricing.reasoningOutputPerMillion ?? modelPricing.outputPerMillion;
|
|
551
|
-
const cost = billableInput / 1e6 * modelPricing.inputPerMillion + usage.cachedInputTokens / 1e6 * cachedRate + cacheCreationInput5m / 1e6 * cacheCreation5mRate + cacheCreationInput1h / 1e6 * cacheCreation1hRate + unclassifiedCacheCreationInput / 1e6 * cacheCreationRate + usage.outputTokens / 1e6 * modelPricing.outputPerMillion + usage.reasoningTokens / 1e6 * reasoningRate;
|
|
552
|
-
return
|
|
731
|
+
const cost = billableInput / 1e6 * modelPricing.inputPerMillion + usage.cachedInputTokens / 1e6 * cachedRate + cacheCreationInput5m / 1e6 * cacheCreation5mRate + cacheCreationInput1h / 1e6 * cacheCreation1hRate + cacheCreationInput30m / 1e6 * cacheCreation30mRate + cacheCreationInput12h / 1e6 * cacheCreation12hRate + unclassifiedCacheCreationInput / 1e6 * cacheCreationRate + usage.outputTokens / 1e6 * modelPricing.outputPerMillion + usage.reasoningTokens / 1e6 * reasoningRate;
|
|
732
|
+
return cost;
|
|
553
733
|
}
|
|
554
734
|
function firstPositiveRate(...rates) {
|
|
555
735
|
return rates.find((rate) => typeof rate === "number" && Number.isFinite(rate) && rate > 0) ?? 0;
|
|
556
736
|
}
|
|
557
737
|
|
|
558
738
|
// packages/core/src/cache.ts
|
|
559
|
-
var activeCacheVersion = "parse-
|
|
739
|
+
var activeCacheVersion = "parse-v2";
|
|
560
740
|
var preparedCacheRoot;
|
|
561
741
|
function parseCacheRoot() {
|
|
562
742
|
return path2.join(repospendHome(), "cache", activeCacheVersion);
|
|
@@ -820,7 +1000,7 @@ function filterUsage(sessions, filters) {
|
|
|
820
1000
|
if (!matchesAny(session.sourceClient, filters.source)) return false;
|
|
821
1001
|
if (!matchesAny(session.sourceApp, filters.sourceApp)) return false;
|
|
822
1002
|
if (!matchesRepo(session, filters.repo)) return false;
|
|
823
|
-
if (!matchesModel(session
|
|
1003
|
+
if (!matchesModel(session, filters.model)) return false;
|
|
824
1004
|
const startedAt = timestamp(session.startedAt);
|
|
825
1005
|
const hasDateFilter = Boolean(filters.from || filters.to);
|
|
826
1006
|
if (hasDateFilter && startedAt === void 0) return false;
|
|
@@ -829,7 +1009,38 @@ function filterUsage(sessions, filters) {
|
|
|
829
1009
|
if (from !== void 0 && startedAt !== void 0 && startedAt < from) return false;
|
|
830
1010
|
if (to !== void 0 && startedAt !== void 0 && startedAt > to) return false;
|
|
831
1011
|
return true;
|
|
832
|
-
});
|
|
1012
|
+
}).map((session) => projectModelFilter(session, filters.model));
|
|
1013
|
+
}
|
|
1014
|
+
function projectModelFilter(session, filter) {
|
|
1015
|
+
const values = filterValues(filter);
|
|
1016
|
+
if (!values.length || !session.modelUsage?.length) return session;
|
|
1017
|
+
const slices = session.modelUsage.filter((entry) => modelValueMatches(entry.model, values));
|
|
1018
|
+
const reasoningTokens = sumSlices(slices, "reasoningTokens");
|
|
1019
|
+
return {
|
|
1020
|
+
...session,
|
|
1021
|
+
// The representative model must belong to the selected usage: keep the original
|
|
1022
|
+
// model when it is selected, otherwise use the first matching slice's model.
|
|
1023
|
+
model: session.model !== void 0 && slices.some((entry) => entry.model === session.model) ? session.model : slices[0]?.model,
|
|
1024
|
+
modelUsage: slices,
|
|
1025
|
+
inputTokens: sumSlices(slices, "inputTokens"),
|
|
1026
|
+
cachedInputTokens: sumSlices(slices, "cachedInputTokens"),
|
|
1027
|
+
cacheCreationInputTokens: sumSlices(slices, "cacheCreationInputTokens"),
|
|
1028
|
+
cacheCreationInputTokens5m: sumSlices(slices, "cacheCreationInputTokens5m"),
|
|
1029
|
+
cacheCreationInputTokens1h: sumSlices(slices, "cacheCreationInputTokens1h"),
|
|
1030
|
+
cacheCreationInputTokens30m: sumSlices(slices, "cacheCreationInputTokens30m"),
|
|
1031
|
+
cacheCreationInputTokens12h: sumSlices(slices, "cacheCreationInputTokens12h"),
|
|
1032
|
+
outputTokens: sumSlices(slices, "outputTokens"),
|
|
1033
|
+
reasoningTokens,
|
|
1034
|
+
reasoningOutputTokens: reasoningTokens,
|
|
1035
|
+
totalTokens: sumSlices(slices, "totalTokens"),
|
|
1036
|
+
estimatedCostUsd: slices.every((entry) => entry.estimatedCostUsd !== void 0) ? slices.reduce((total, entry) => total + (entry.estimatedCostUsd ?? 0), 0) : void 0
|
|
1037
|
+
};
|
|
1038
|
+
}
|
|
1039
|
+
function modelValueMatches(model, values) {
|
|
1040
|
+
return values.includes(model ?? "unknown-model");
|
|
1041
|
+
}
|
|
1042
|
+
function sumSlices(slices, key) {
|
|
1043
|
+
return slices.reduce((total, entry) => total + (entry[key] ?? 0), 0);
|
|
833
1044
|
}
|
|
834
1045
|
function matchesAny(value, filter) {
|
|
835
1046
|
const values = filterValues(filter);
|
|
@@ -841,10 +1052,11 @@ function matchesRepo(session, filter) {
|
|
|
841
1052
|
if (!values.length) return true;
|
|
842
1053
|
return values.includes(session.repoName) || values.includes(session.repoRoot);
|
|
843
1054
|
}
|
|
844
|
-
function matchesModel(
|
|
1055
|
+
function matchesModel(session, filter) {
|
|
845
1056
|
const values = filterValues(filter);
|
|
846
1057
|
if (!values.length) return true;
|
|
847
|
-
|
|
1058
|
+
const models = session.modelUsage?.length ? session.modelUsage.map((entry) => entry.model) : [session.model];
|
|
1059
|
+
return models.some((model) => model !== void 0 ? values.includes(model) : values.includes("unknown-model"));
|
|
848
1060
|
}
|
|
849
1061
|
function filterValues(filter) {
|
|
850
1062
|
if (!filter) return [];
|
|
@@ -1841,28 +2053,35 @@ function dateValue(value) {
|
|
|
1841
2053
|
}
|
|
1842
2054
|
|
|
1843
2055
|
// packages/core/src/sources/codex-token.ts
|
|
1844
|
-
function aggregateTokens(records) {
|
|
2056
|
+
function aggregateTokens(records, modelForRecord) {
|
|
1845
2057
|
const accumulator = emptyTokenAccumulator();
|
|
1846
2058
|
for (const record of records) {
|
|
1847
|
-
collectRecordTokens(accumulator, record);
|
|
2059
|
+
collectRecordTokens(accumulator, record, modelForRecord?.(record));
|
|
1848
2060
|
}
|
|
1849
2061
|
finalizeTokenAccumulator(accumulator);
|
|
1850
2062
|
return {
|
|
1851
2063
|
inputTokens: accumulator.inputTokens,
|
|
1852
2064
|
cachedInputTokens: accumulator.cachedInputTokens,
|
|
2065
|
+
cacheCreationInputTokens: accumulator.cacheCreationInputTokens || void 0,
|
|
2066
|
+
cacheCreationInputTokens30m: accumulator.cacheCreationInputTokens30m || void 0,
|
|
2067
|
+
cacheCreationInputTokens12h: accumulator.cacheCreationInputTokens12h || void 0,
|
|
1853
2068
|
outputTokens: accumulator.outputTokens,
|
|
1854
2069
|
reasoningTokens: accumulator.reasoningTokens,
|
|
1855
2070
|
totalTokens: accumulator.totalTokens,
|
|
1856
2071
|
tokenAggregationMethod: accumulator.tokenAggregationMethod,
|
|
1857
2072
|
tokenConfidence: accumulator.tokenConfidence,
|
|
1858
2073
|
tokenSnapshotCount: accumulator.tokenSnapshotCount,
|
|
1859
|
-
warnings: accumulator.warnings
|
|
2074
|
+
warnings: accumulator.warnings,
|
|
2075
|
+
requestUsages: accumulator.requestUsages
|
|
1860
2076
|
};
|
|
1861
2077
|
}
|
|
1862
2078
|
function emptyTokenAccumulator() {
|
|
1863
2079
|
return {
|
|
1864
2080
|
inputTokens: 0,
|
|
1865
2081
|
cachedInputTokens: 0,
|
|
2082
|
+
cacheCreationInputTokens: 0,
|
|
2083
|
+
cacheCreationInputTokens30m: 0,
|
|
2084
|
+
cacheCreationInputTokens12h: 0,
|
|
1866
2085
|
outputTokens: 0,
|
|
1867
2086
|
reasoningTokens: 0,
|
|
1868
2087
|
totalTokens: 0,
|
|
@@ -1871,12 +2090,14 @@ function emptyTokenAccumulator() {
|
|
|
1871
2090
|
tokenSnapshotCount: 0,
|
|
1872
2091
|
warnings: [],
|
|
1873
2092
|
cumulativeTokenSnapshots: [],
|
|
1874
|
-
directTokenUsages: []
|
|
2093
|
+
directTokenUsages: [],
|
|
2094
|
+
requestUsages: []
|
|
1875
2095
|
};
|
|
1876
2096
|
}
|
|
1877
|
-
function collectRecordTokens(accumulator, record) {
|
|
2097
|
+
function collectRecordTokens(accumulator, record, model) {
|
|
1878
2098
|
const usage = extractRecordTokens(record);
|
|
1879
2099
|
if (!usage) return;
|
|
2100
|
+
usage.model = model;
|
|
1880
2101
|
accumulator.tokenSnapshotCount += 1;
|
|
1881
2102
|
if (usage.cumulative) {
|
|
1882
2103
|
accumulator.cumulativeTokenSnapshots.push(usage);
|
|
@@ -1893,6 +2114,7 @@ function finalizeTokenAccumulator(accumulator) {
|
|
|
1893
2114
|
applyFinalUsage(accumulator, finalSnapshot);
|
|
1894
2115
|
accumulator.tokenAggregationMethod = "final_snapshot";
|
|
1895
2116
|
accumulator.tokenConfidence = "high";
|
|
2117
|
+
accumulator.requestUsages = [{ ...requestUsageFrom(finalSnapshot, void 0), model: void 0 }];
|
|
1896
2118
|
if (accumulator.directTokenUsages.length > 0) accumulator.warnings.push("token_direct_usage_ignored_after_cumulative_snapshot");
|
|
1897
2119
|
return;
|
|
1898
2120
|
}
|
|
@@ -1903,15 +2125,19 @@ function finalizeTokenAccumulator(accumulator) {
|
|
|
1903
2125
|
(sum2, usage) => ({
|
|
1904
2126
|
inputTokens: sum2.inputTokens + usage.inputTokens,
|
|
1905
2127
|
cachedInputTokens: sum2.cachedInputTokens + usage.cachedInputTokens,
|
|
2128
|
+
cacheCreationInputTokens: sum2.cacheCreationInputTokens + usage.cacheCreationInputTokens,
|
|
2129
|
+
cacheCreationInputTokens30m: sum2.cacheCreationInputTokens30m + usage.cacheCreationInputTokens30m,
|
|
2130
|
+
cacheCreationInputTokens12h: sum2.cacheCreationInputTokens12h + usage.cacheCreationInputTokens12h,
|
|
1906
2131
|
outputTokens: sum2.outputTokens + usage.outputTokens,
|
|
1907
2132
|
reasoningTokens: sum2.reasoningTokens + usage.reasoningTokens,
|
|
1908
2133
|
totalTokens: sum2.totalTokens + usage.totalTokens
|
|
1909
2134
|
}),
|
|
1910
|
-
{ inputTokens: 0, cachedInputTokens: 0, outputTokens: 0, reasoningTokens: 0, totalTokens: 0 }
|
|
2135
|
+
{ inputTokens: 0, cachedInputTokens: 0, cacheCreationInputTokens: 0, cacheCreationInputTokens30m: 0, cacheCreationInputTokens12h: 0, outputTokens: 0, reasoningTokens: 0, totalTokens: 0 }
|
|
1911
2136
|
);
|
|
1912
2137
|
applyFinalUsage(accumulator, total);
|
|
1913
2138
|
accumulator.tokenAggregationMethod = "delta_sum";
|
|
1914
2139
|
accumulator.tokenConfidence = "high";
|
|
2140
|
+
accumulator.requestUsages = filteredDirect.map((usage) => requestUsageFrom(usage, usage.inputTokens));
|
|
1915
2141
|
const skipped = direct.length - filteredDirect.length;
|
|
1916
2142
|
if (skipped > 0) accumulator.warnings.push("duplicate_or_stale_token_snapshots_skipped");
|
|
1917
2143
|
return;
|
|
@@ -1919,6 +2145,20 @@ function finalizeTokenAccumulator(accumulator) {
|
|
|
1919
2145
|
accumulator.tokenAggregationMethod = "unknown";
|
|
1920
2146
|
accumulator.tokenConfidence = "low";
|
|
1921
2147
|
}
|
|
2148
|
+
function requestUsageFrom(usage, requestInputTokens) {
|
|
2149
|
+
return {
|
|
2150
|
+
model: "model" in usage ? usage.model : void 0,
|
|
2151
|
+
requestInputTokens,
|
|
2152
|
+
inputTokens: usage.inputTokens,
|
|
2153
|
+
cachedInputTokens: usage.cachedInputTokens,
|
|
2154
|
+
cacheCreationInputTokens: usage.cacheCreationInputTokens,
|
|
2155
|
+
cacheCreationInputTokens30m: usage.cacheCreationInputTokens30m,
|
|
2156
|
+
cacheCreationInputTokens12h: usage.cacheCreationInputTokens12h,
|
|
2157
|
+
outputTokens: usage.outputTokens,
|
|
2158
|
+
reasoningTokens: usage.reasoningTokens,
|
|
2159
|
+
totalTokens: usage.totalTokens
|
|
2160
|
+
};
|
|
2161
|
+
}
|
|
1922
2162
|
function extractRecordTokens(record) {
|
|
1923
2163
|
if (!record || typeof record !== "object") return void 0;
|
|
1924
2164
|
const object = record;
|
|
@@ -1954,10 +2194,10 @@ function filterDuplicateDirectUsages(usages) {
|
|
|
1954
2194
|
return filtered;
|
|
1955
2195
|
}
|
|
1956
2196
|
function sameTokenUsage(a, b) {
|
|
1957
|
-
return a.inputTokens === b.inputTokens && a.cachedInputTokens === b.cachedInputTokens && a.outputTokens === b.outputTokens && a.reasoningTokens === b.reasoningTokens && a.totalTokens === b.totalTokens;
|
|
2197
|
+
return a.inputTokens === b.inputTokens && a.cachedInputTokens === b.cachedInputTokens && a.cacheCreationInputTokens === b.cacheCreationInputTokens && a.cacheCreationInputTokens30m === b.cacheCreationInputTokens30m && a.cacheCreationInputTokens12h === b.cacheCreationInputTokens12h && a.outputTokens === b.outputTokens && a.reasoningTokens === b.reasoningTokens && a.totalTokens === b.totalTokens;
|
|
1958
2198
|
}
|
|
1959
2199
|
function isMonotonicFrom(current, previous) {
|
|
1960
|
-
return current.inputTokens >= previous.inputTokens && current.cachedInputTokens >= previous.cachedInputTokens && current.outputTokens >= previous.outputTokens && current.reasoningTokens >= previous.reasoningTokens;
|
|
2200
|
+
return current.inputTokens >= previous.inputTokens && current.cachedInputTokens >= previous.cachedInputTokens && current.cacheCreationInputTokens >= previous.cacheCreationInputTokens && current.outputTokens >= previous.outputTokens && current.reasoningTokens >= previous.reasoningTokens;
|
|
1961
2201
|
}
|
|
1962
2202
|
function looksLikeStaleRegression(current, previous, last) {
|
|
1963
2203
|
const previousTotal = tokenBucketTotal(previous);
|
|
@@ -1973,22 +2213,34 @@ function stripSnapshot(usage) {
|
|
|
1973
2213
|
return {
|
|
1974
2214
|
inputTokens: usage.inputTokens,
|
|
1975
2215
|
cachedInputTokens: usage.cachedInputTokens,
|
|
2216
|
+
cacheCreationInputTokens: usage.cacheCreationInputTokens,
|
|
2217
|
+
cacheCreationInputTokens30m: usage.cacheCreationInputTokens30m,
|
|
2218
|
+
cacheCreationInputTokens12h: usage.cacheCreationInputTokens12h,
|
|
1976
2219
|
outputTokens: usage.outputTokens,
|
|
1977
2220
|
reasoningTokens: usage.reasoningTokens,
|
|
1978
2221
|
totalTokens: usage.totalTokens
|
|
1979
2222
|
};
|
|
1980
2223
|
}
|
|
1981
2224
|
function tokenUsageFromObject(target, cumulative) {
|
|
2225
|
+
const inputDetails = firstObject2(target.input_tokens_details, target.inputTokensDetails, target.prompt_tokens_details, target.promptTokensDetails);
|
|
2226
|
+
const ttlDetails = firstObject2(target.cache_write_tokens_details, target.cacheWriteTokensDetails, inputDetails?.cache_write_tokens_details, inputDetails?.cacheWriteTokensDetails);
|
|
1982
2227
|
const input = numberFromKeys(target, ["input_tokens", "inputTokens", "prompt_tokens", "promptTokens"]);
|
|
1983
|
-
const cached = numberFromKeys(target, ["cached_input_tokens", "cachedInputTokens", "cached_tokens", "cachedTokens", "cache_read_input_tokens"]);
|
|
2228
|
+
const cached = numberFromKeys(target, ["cached_input_tokens", "cachedInputTokens", "cached_tokens", "cachedTokens", "cache_read_input_tokens"]) || (inputDetails ? numberFromKeys(inputDetails, ["cached_tokens", "cachedTokens"]) : 0);
|
|
1984
2229
|
const outputWithReasoning = numberFromKeys(target, ["output_tokens", "outputTokens", "completion_tokens", "completionTokens"]);
|
|
1985
|
-
const reasoning = numberFromKeys(target, ["reasoning_tokens", "reasoningTokens", "reasoning_output_tokens", "reasoningOutputTokens"]);
|
|
2230
|
+
const reasoning = numberFromKeys(target, ["reasoning_tokens", "reasoningTokens", "reasoning_output_tokens", "reasoningOutputTokens"]) || (firstObject2(target.output_tokens_details, target.outputTokensDetails) ? numberFromKeys(firstObject2(target.output_tokens_details, target.outputTokensDetails), ["reasoning_tokens", "reasoningTokens"]) : 0);
|
|
1986
2231
|
const output = Math.max(outputWithReasoning - reasoning, 0);
|
|
1987
2232
|
const total = numberFromKeys(target, ["total_tokens", "totalTokens", "tokens_used", "tokensUsed"]);
|
|
1988
|
-
|
|
2233
|
+
const cacheWrite30m = ttlDetails ? numberFromKeys(ttlDetails, ["ttl_30m_tokens", "ttl30mTokens", "ttl_30m"]) : 0;
|
|
2234
|
+
const cacheWrite12h = ttlDetails ? numberFromKeys(ttlDetails, ["ttl_12h_tokens", "ttl12hTokens", "ttl_12h"]) : 0;
|
|
2235
|
+
const cacheWriteDeclared = numberFromKeys(target, ["cache_creation_input_tokens", "cacheCreationInputTokens", "cache_write_input_tokens", "cacheWriteInputTokens", "cache_write_tokens", "cacheWriteTokens"]) || (inputDetails ? numberFromKeys(inputDetails, ["cache_write_tokens", "cacheWriteTokens", "cache_creation_input_tokens", "cacheCreationInputTokens"]) : 0);
|
|
2236
|
+
const cacheCreation = Math.max(cacheWriteDeclared, cacheWrite30m + cacheWrite12h);
|
|
2237
|
+
if (input + cached + output + reasoning + total + cacheCreation === 0) return void 0;
|
|
1989
2238
|
return {
|
|
1990
2239
|
inputTokens: input,
|
|
1991
2240
|
cachedInputTokens: cached,
|
|
2241
|
+
cacheCreationInputTokens: cacheCreation,
|
|
2242
|
+
cacheCreationInputTokens30m: cacheWrite30m,
|
|
2243
|
+
cacheCreationInputTokens12h: cacheWrite12h,
|
|
1992
2244
|
outputTokens: output,
|
|
1993
2245
|
reasoningTokens: reasoning,
|
|
1994
2246
|
totalTokens: total || input + output + reasoning,
|
|
@@ -1998,6 +2250,9 @@ function tokenUsageFromObject(target, cumulative) {
|
|
|
1998
2250
|
function applyFinalUsage(accumulator, usage) {
|
|
1999
2251
|
accumulator.inputTokens = usage.inputTokens;
|
|
2000
2252
|
accumulator.cachedInputTokens = usage.cachedInputTokens;
|
|
2253
|
+
accumulator.cacheCreationInputTokens = usage.cacheCreationInputTokens;
|
|
2254
|
+
accumulator.cacheCreationInputTokens30m = usage.cacheCreationInputTokens30m;
|
|
2255
|
+
accumulator.cacheCreationInputTokens12h = usage.cacheCreationInputTokens12h;
|
|
2001
2256
|
accumulator.outputTokens = usage.outputTokens;
|
|
2002
2257
|
accumulator.reasoningTokens = usage.reasoningTokens;
|
|
2003
2258
|
accumulator.totalTokens = usage.totalTokens || usage.inputTokens + usage.outputTokens + usage.reasoningTokens;
|
|
@@ -2018,7 +2273,7 @@ function numberFromKeys(object, keys) {
|
|
|
2018
2273
|
}
|
|
2019
2274
|
|
|
2020
2275
|
// packages/core/src/sources/codex.ts
|
|
2021
|
-
var codexSessionCacheVersion = "codex-session-
|
|
2276
|
+
var codexSessionCacheVersion = "codex-session-v2";
|
|
2022
2277
|
var nativeSqliteRebuildAttempted = false;
|
|
2023
2278
|
function scanCodex(options) {
|
|
2024
2279
|
const codexHome = options.codexHome ?? path5.join(os3.homedir(), ".codex");
|
|
@@ -2249,6 +2504,9 @@ function threadToUsage(row, index, sessionFiles, options, importedModels = /* @_
|
|
|
2249
2504
|
provider,
|
|
2250
2505
|
inputTokens,
|
|
2251
2506
|
cachedInputTokens,
|
|
2507
|
+
cacheCreationInputTokens: sessionBreakdown.cacheCreationInputTokens,
|
|
2508
|
+
cacheCreationInputTokens30m: sessionBreakdown.cacheCreationInputTokens30m,
|
|
2509
|
+
cacheCreationInputTokens12h: sessionBreakdown.cacheCreationInputTokens12h,
|
|
2252
2510
|
outputTokens,
|
|
2253
2511
|
reasoningTokens,
|
|
2254
2512
|
reasoningOutputTokens: reasoningTokens,
|
|
@@ -2288,12 +2546,14 @@ function threadToUsage(row, index, sessionFiles, options, importedModels = /* @_
|
|
|
2288
2546
|
sessionOutcome: inferOutcome(sessionBreakdown),
|
|
2289
2547
|
compaction: sessionBreakdown.compaction
|
|
2290
2548
|
};
|
|
2291
|
-
const
|
|
2292
|
-
const
|
|
2549
|
+
const estimate = hasPricedBreakdown ? sessionEstimate(sessionBreakdown.requestUsages, model, options.pricing) : void 0;
|
|
2550
|
+
const cost = estimate?.result.costUsd;
|
|
2551
|
+
const costWarnings = [...warnings, ...estimate?.result.assumptions ?? []];
|
|
2293
2552
|
return {
|
|
2294
2553
|
...usage,
|
|
2295
2554
|
estimatedCostUsd: cost,
|
|
2296
|
-
|
|
2555
|
+
modelUsage: estimate?.modelUsage,
|
|
2556
|
+
warnings: cost === void 0 ? [.../* @__PURE__ */ new Set([...hasPricedBreakdown ? costWarnings : [...costWarnings, "missing_token_breakdown"], "unknown_pricing", ...estimate?.result.warnings ?? []])] : [...new Set(costWarnings)]
|
|
2297
2557
|
};
|
|
2298
2558
|
}
|
|
2299
2559
|
function readStandaloneSessions(sessionFiles, options) {
|
|
@@ -2323,6 +2583,9 @@ function readStandaloneSessions(sessionFiles, options) {
|
|
|
2323
2583
|
provider,
|
|
2324
2584
|
inputTokens: breakdown.inputTokens,
|
|
2325
2585
|
cachedInputTokens: breakdown.cachedInputTokens,
|
|
2586
|
+
cacheCreationInputTokens: breakdown.cacheCreationInputTokens,
|
|
2587
|
+
cacheCreationInputTokens30m: breakdown.cacheCreationInputTokens30m,
|
|
2588
|
+
cacheCreationInputTokens12h: breakdown.cacheCreationInputTokens12h,
|
|
2326
2589
|
outputTokens: breakdown.outputTokens,
|
|
2327
2590
|
reasoningTokens: breakdown.reasoningTokens,
|
|
2328
2591
|
reasoningOutputTokens: breakdown.reasoningTokens,
|
|
@@ -2363,14 +2626,53 @@ function readStandaloneSessions(sessionFiles, options) {
|
|
|
2363
2626
|
compaction: breakdown.compaction
|
|
2364
2627
|
};
|
|
2365
2628
|
const hasPricedBreakdown = breakdown.inputTokens + breakdown.cachedInputTokens + breakdown.outputTokens + breakdown.reasoningTokens > 0;
|
|
2366
|
-
const
|
|
2629
|
+
const estimate = hasPricedBreakdown ? sessionEstimate(breakdown.requestUsages, model, options.pricing) : void 0;
|
|
2630
|
+
const cost = estimate?.result.costUsd;
|
|
2367
2631
|
return {
|
|
2368
2632
|
...usage,
|
|
2369
2633
|
estimatedCostUsd: cost,
|
|
2370
|
-
|
|
2634
|
+
modelUsage: estimate?.modelUsage,
|
|
2635
|
+
warnings: cost === void 0 ? [.../* @__PURE__ */ new Set([...usage.warnings, hasPricedBreakdown ? "unknown_pricing" : "missing_token_breakdown", ...estimate?.result.warnings ?? [], ...estimate?.result.assumptions ?? []])] : [.../* @__PURE__ */ new Set([...usage.warnings, ...estimate?.result.assumptions ?? []])]
|
|
2371
2636
|
};
|
|
2372
2637
|
});
|
|
2373
2638
|
}
|
|
2639
|
+
function sessionEstimate(requestUsages, fallbackModel, pricing) {
|
|
2640
|
+
const requests = requestUsages.map((request) => ({ ...request, model: request.model ?? fallbackModel }));
|
|
2641
|
+
const result = calculateRequestCosts(requests, pricing);
|
|
2642
|
+
return { result, modelUsage: buildSessionModelUsage(requests, result.requestCosts) };
|
|
2643
|
+
}
|
|
2644
|
+
function buildSessionModelUsage(requests, requestCosts) {
|
|
2645
|
+
if (!requests.length) return void 0;
|
|
2646
|
+
const groups = /* @__PURE__ */ new Map();
|
|
2647
|
+
requests.forEach((request, index) => {
|
|
2648
|
+
const entry = groups.get(request.model) ?? {
|
|
2649
|
+
model: request.model,
|
|
2650
|
+
requestCount: 0,
|
|
2651
|
+
inputTokens: 0,
|
|
2652
|
+
cachedInputTokens: 0,
|
|
2653
|
+
cacheCreationInputTokens: 0,
|
|
2654
|
+
cacheCreationInputTokens30m: 0,
|
|
2655
|
+
cacheCreationInputTokens12h: 0,
|
|
2656
|
+
outputTokens: 0,
|
|
2657
|
+
reasoningTokens: 0,
|
|
2658
|
+
totalTokens: 0,
|
|
2659
|
+
estimatedCostUsd: 0
|
|
2660
|
+
};
|
|
2661
|
+
entry.requestCount += 1;
|
|
2662
|
+
entry.inputTokens += request.inputTokens;
|
|
2663
|
+
entry.cachedInputTokens += request.cachedInputTokens;
|
|
2664
|
+
entry.cacheCreationInputTokens += request.cacheCreationInputTokens;
|
|
2665
|
+
entry.cacheCreationInputTokens30m = (entry.cacheCreationInputTokens30m ?? 0) + request.cacheCreationInputTokens30m;
|
|
2666
|
+
entry.cacheCreationInputTokens12h = (entry.cacheCreationInputTokens12h ?? 0) + request.cacheCreationInputTokens12h;
|
|
2667
|
+
entry.outputTokens += request.outputTokens;
|
|
2668
|
+
entry.reasoningTokens += request.reasoningTokens;
|
|
2669
|
+
entry.totalTokens += request.totalTokens;
|
|
2670
|
+
const requestCost = requestCosts[index];
|
|
2671
|
+
entry.estimatedCostUsd = requestCost === void 0 || entry.estimatedCostUsd === void 0 ? void 0 : Number((entry.estimatedCostUsd + requestCost).toFixed(6));
|
|
2672
|
+
groups.set(request.model, entry);
|
|
2673
|
+
});
|
|
2674
|
+
return [...groups.values()];
|
|
2675
|
+
}
|
|
2374
2676
|
function hasImportedSession(sessions, candidate) {
|
|
2375
2677
|
const candidatePath = comparableSourcePath(candidate.sourcePath);
|
|
2376
2678
|
return sessions.some((session) => {
|
|
@@ -2515,7 +2817,8 @@ function readSessionBreakdown(filePath) {
|
|
|
2515
2817
|
applyCompactionEvent(breakdown, record, lastUserMessageText);
|
|
2516
2818
|
}
|
|
2517
2819
|
breakdown.messageCount = eventMessageCount || responseMessageCount;
|
|
2518
|
-
const
|
|
2820
|
+
const modelByRecord = recordModels(records);
|
|
2821
|
+
const tokenAggregation = aggregateTokens(records, (record) => modelByRecord.get(record));
|
|
2519
2822
|
Object.assign(breakdown, {
|
|
2520
2823
|
...tokenAggregation,
|
|
2521
2824
|
warnings: [...breakdown.warnings, ...tokenAggregation.warnings]
|
|
@@ -2530,6 +2833,22 @@ function readSessionBreakdown(filePath) {
|
|
|
2530
2833
|
if (cacheKey) writeJsonCache("codex-session", cacheKey, breakdown);
|
|
2531
2834
|
return breakdown;
|
|
2532
2835
|
}
|
|
2836
|
+
function recordModels(records) {
|
|
2837
|
+
const models = /* @__PURE__ */ new Map();
|
|
2838
|
+
let current;
|
|
2839
|
+
for (const record of records) {
|
|
2840
|
+
const declared = declaredRecordModel(record);
|
|
2841
|
+
if (declared) current = declared;
|
|
2842
|
+
models.set(record, current);
|
|
2843
|
+
}
|
|
2844
|
+
return models;
|
|
2845
|
+
}
|
|
2846
|
+
function declaredRecordModel(record) {
|
|
2847
|
+
const object = asRecord(record);
|
|
2848
|
+
const payload = firstObject3(object?.payload) ?? object;
|
|
2849
|
+
const collaborationSettings = firstObject3(firstObject3(payload?.collaboration_mode)?.settings);
|
|
2850
|
+
return stringValue2(payload?.model) ?? stringValue2(collaborationSettings?.model);
|
|
2851
|
+
}
|
|
2533
2852
|
function sessionBreakdownCacheKey(filePath) {
|
|
2534
2853
|
try {
|
|
2535
2854
|
const stat = fs5.statSync(filePath);
|
|
@@ -2614,6 +2933,7 @@ function emptyBreakdown() {
|
|
|
2614
2933
|
outputTokens: 0,
|
|
2615
2934
|
reasoningTokens: 0,
|
|
2616
2935
|
totalTokens: 0,
|
|
2936
|
+
requestUsages: [],
|
|
2617
2937
|
tokenAggregationMethod: "unknown",
|
|
2618
2938
|
tokenConfidence: "low",
|
|
2619
2939
|
tokenSnapshotCount: 0,
|
|
@@ -3053,21 +3373,30 @@ function claudeBreakdownToUsage(file, index, options, breakdown, idSuffix = "")
|
|
|
3053
3373
|
promptTimeline: breakdown.promptTimeline,
|
|
3054
3374
|
sessionOutcome: inferOutcome2(breakdown)
|
|
3055
3375
|
};
|
|
3056
|
-
const
|
|
3057
|
-
for (const event of breakdown.usageEvents) {
|
|
3376
|
+
const requests = breakdown.usageEvents.map((event) => {
|
|
3058
3377
|
const model = event.model ?? breakdown.model;
|
|
3059
3378
|
const normalized = model ? normalizePricingModelId(model) : void 0;
|
|
3060
3379
|
const pricingModel = normalized && normalizedServiceTier(event.usage.speed) === "fast" && !normalized.includes("fast-mode") && !/^claude-opus-4-6(?:$|-)/.test(normalized) ? `${normalized}-fast-mode` : model;
|
|
3061
|
-
const totals =
|
|
3380
|
+
const totals = emptyBreakdown2();
|
|
3062
3381
|
applyUsage(totals, event.usage);
|
|
3063
|
-
|
|
3064
|
-
|
|
3065
|
-
|
|
3066
|
-
|
|
3382
|
+
return {
|
|
3383
|
+
model: pricingModel,
|
|
3384
|
+
inputTokens: totals.inputTokens,
|
|
3385
|
+
cachedInputTokens: totals.cachedInputTokens,
|
|
3386
|
+
cacheCreationInputTokens: totals.cacheCreationInputTokens,
|
|
3387
|
+
cacheCreationInputTokens5m: totals.cacheCreationInputTokens5m,
|
|
3388
|
+
cacheCreationInputTokens1h: totals.cacheCreationInputTokens1h,
|
|
3389
|
+
outputTokens: totals.outputTokens,
|
|
3390
|
+
reasoningTokens: 0,
|
|
3391
|
+
requestInputTokens: hasInputTokenCount(event.usage) ? totals.inputTokens : void 0
|
|
3392
|
+
};
|
|
3393
|
+
});
|
|
3394
|
+
const estimate = hasTokenBreakdown ? calculateRequestCosts(requests, options.pricing) : void 0;
|
|
3395
|
+
const cost = estimate?.costUsd;
|
|
3067
3396
|
return {
|
|
3068
3397
|
...usage,
|
|
3069
3398
|
estimatedCostUsd: cost,
|
|
3070
|
-
warnings: cost === void 0 ? [...usage.warnings, "unknown_pricing", ...
|
|
3399
|
+
warnings: cost === void 0 ? [.../* @__PURE__ */ new Set([...usage.warnings, "unknown_pricing", ...estimate?.warnings ?? [], ...estimate?.assumptions ?? [], ...estimate?.warnings.some((warning) => warning.startsWith("unknown_pricing_model:") && warning.includes("fast-mode")) ? ["unknown_fast_mode_pricing"] : []])] : [.../* @__PURE__ */ new Set([...usage.warnings, ...estimate?.assumptions ?? []])]
|
|
3071
3400
|
};
|
|
3072
3401
|
}
|
|
3073
3402
|
function inferClaudeEntrypointFromPath(filePath) {
|
|
@@ -3169,10 +3498,14 @@ function usageDedupKey(record, message) {
|
|
|
3169
3498
|
function occurrenceKey(filePath, recordIndex) {
|
|
3170
3499
|
return `${path6.resolve(filePath)}:${recordIndex}`;
|
|
3171
3500
|
}
|
|
3501
|
+
function hasInputTokenCount(usage) {
|
|
3502
|
+
const value = usage.input_tokens;
|
|
3503
|
+
return (typeof value === "number" || typeof value === "string" && value.trim() !== "") && Number.isFinite(Number(value)) && Number(value) >= 0;
|
|
3504
|
+
}
|
|
3172
3505
|
function cloneUsage(usage) {
|
|
3173
3506
|
const cacheCreation = firstObject4(usage.cache_creation);
|
|
3174
3507
|
return {
|
|
3175
|
-
input_tokens: numberValue(usage.input_tokens),
|
|
3508
|
+
input_tokens: hasInputTokenCount(usage) ? numberValue(usage.input_tokens) : void 0,
|
|
3176
3509
|
cache_read_input_tokens: numberValue(usage.cache_read_input_tokens),
|
|
3177
3510
|
cache_creation_input_tokens: numberValue(usage.cache_creation_input_tokens),
|
|
3178
3511
|
cache_creation: cacheCreation ? {
|
|
@@ -3187,7 +3520,7 @@ function cloneUsage(usage) {
|
|
|
3187
3520
|
function mergeUsageMax(target, usage) {
|
|
3188
3521
|
const targetCacheCreation = firstObject4(target.cache_creation) ?? {};
|
|
3189
3522
|
const usageCacheCreation = firstObject4(usage.cache_creation) ?? {};
|
|
3190
|
-
target.input_tokens = Math.max(numberValue(target.input_tokens), numberValue(usage.input_tokens));
|
|
3523
|
+
if (hasInputTokenCount(target) || hasInputTokenCount(usage)) target.input_tokens = Math.max(numberValue(target.input_tokens), numberValue(usage.input_tokens));
|
|
3191
3524
|
target.cache_read_input_tokens = Math.max(numberValue(target.cache_read_input_tokens), numberValue(usage.cache_read_input_tokens));
|
|
3192
3525
|
target.cache_creation_input_tokens = Math.max(numberValue(target.cache_creation_input_tokens), numberValue(usage.cache_creation_input_tokens));
|
|
3193
3526
|
target.cache_creation = {
|
|
@@ -3770,6 +4103,8 @@ function usageCandidateFromRecord(record, index, fallbackTimestamp, traceContext
|
|
|
3770
4103
|
const output = numberValue2(attributes["gen_ai.usage.output_tokens"]);
|
|
3771
4104
|
const cached = numberValue2(attributes["gen_ai.usage.cache_read.input_tokens"]);
|
|
3772
4105
|
const cacheCreation = firstPositiveNumber(attributes, ["gen_ai.usage.cache_write.input_tokens", "gen_ai.usage.cache_creation.input_tokens"]);
|
|
4106
|
+
const rawInput = attributes["gen_ai.usage.input_tokens"];
|
|
4107
|
+
const requestInputTokens = typeof rawInput === "number" && Number.isFinite(rawInput) && rawInput >= cached + cacheCreation ? rawInput : void 0;
|
|
3773
4108
|
const reasoning = firstPositiveNumber(attributes, ["gen_ai.usage.reasoning.output_tokens", "gen_ai.usage.reasoning_tokens"]);
|
|
3774
4109
|
const total = firstPositiveNumber(attributes, ["gen_ai.usage.total_tokens", "gen_ai.usage.total.token_count"]);
|
|
3775
4110
|
const outputWithFallback = output || outputTokensFromTotal(total, input, reasoning);
|
|
@@ -3786,6 +4121,7 @@ function usageCandidateFromRecord(record, index, fallbackTimestamp, traceContext
|
|
|
3786
4121
|
model: firstNonEmptyAttr(attributes, modelAttrs) ?? traceContext?.model,
|
|
3787
4122
|
timestamp: timestamp3,
|
|
3788
4123
|
durationMs: durationMsFromRecord(record),
|
|
4124
|
+
requestInputTokens,
|
|
3789
4125
|
inputTokens: input,
|
|
3790
4126
|
cachedInputTokens: cached,
|
|
3791
4127
|
cacheCreationInputTokens: cacheCreation,
|
|
@@ -3848,6 +4184,8 @@ function applyCandidate(breakdown, candidate) {
|
|
|
3848
4184
|
breakdown.endedAt = maxDate2(breakdown.endedAt, candidate.timestamp);
|
|
3849
4185
|
applyTokenEvent(breakdown, {
|
|
3850
4186
|
dedupKey: candidate.dedupKey,
|
|
4187
|
+
model: candidate.model,
|
|
4188
|
+
requestInputTokens: candidate.source === "chat_span" || candidate.source === "inference_log" ? candidate.requestInputTokens : void 0,
|
|
3851
4189
|
detailed: candidate.inputTokens > 0 || candidate.cachedInputTokens > 0 || candidate.cacheCreationInputTokens > 0 || candidate.reasoningTokens > 0,
|
|
3852
4190
|
inputTokens: candidate.inputTokens,
|
|
3853
4191
|
cachedInputTokens: candidate.cachedInputTokens,
|
|
@@ -3969,8 +4307,9 @@ function breakdownToUsage(breakdown, index, options) {
|
|
|
3969
4307
|
fileEditCount: breakdown.fileEditCount,
|
|
3970
4308
|
sourceMetadata: { copilot: { sourcePaths: breakdown.sourcePaths, surface: breakdown.surface } }
|
|
3971
4309
|
};
|
|
3972
|
-
const
|
|
3973
|
-
|
|
4310
|
+
const estimate = hasPricedBreakdown ? calculateRequestCosts(breakdown.tokenEvents.map((event) => ({ ...event, model: normalizeCopilotModel(hasKnownModel(event.model) ? event.model : breakdown.model) })), options.pricing) : void 0;
|
|
4311
|
+
const cost = estimate?.costUsd;
|
|
4312
|
+
return { ...usage, estimatedCostUsd: cost, warnings: hasPricedBreakdown && cost === void 0 ? [.../* @__PURE__ */ new Set([...usage.warnings, "unknown_pricing", ...estimate?.warnings ?? [], ...estimate?.assumptions ?? []])] : [.../* @__PURE__ */ new Set([...usage.warnings, ...estimate?.assumptions ?? []])] };
|
|
3974
4313
|
}
|
|
3975
4314
|
function unknownCopilotRepoInfo() {
|
|
3976
4315
|
return {
|
|
@@ -4676,9 +5015,12 @@ function applyCursorRecord(breakdown, record) {
|
|
|
4676
5015
|
}
|
|
4677
5016
|
}
|
|
4678
5017
|
const usage = extractUsage(objects);
|
|
4679
|
-
if (usage.inputTokens + usage.cachedInputTokens + usage.outputTokens + usage.reasoningTokens + usage.totalTokens > 0) {
|
|
5018
|
+
if (usage.inputTokens + usage.cachedInputTokens + usage.outputTokens + usage.reasoningTokens + usage.totalTokens + usage.cacheCreationInputTokens > 0) {
|
|
4680
5019
|
breakdown.inputTokens += usage.inputTokens;
|
|
4681
5020
|
breakdown.cachedInputTokens += usage.cachedInputTokens;
|
|
5021
|
+
breakdown.cacheCreationInputTokens += usage.cacheCreationInputTokens;
|
|
5022
|
+
breakdown.cacheCreationInputTokens30m += usage.cacheCreationInputTokens30m;
|
|
5023
|
+
breakdown.cacheCreationInputTokens12h += usage.cacheCreationInputTokens12h;
|
|
4682
5024
|
breakdown.outputTokens += usage.outputTokens;
|
|
4683
5025
|
breakdown.reasoningTokens += usage.reasoningTokens;
|
|
4684
5026
|
breakdown.totalTokens += usage.totalTokens;
|
|
@@ -4771,9 +5113,16 @@ function textFromUnknown(value) {
|
|
|
4771
5113
|
function extractUsage(objects) {
|
|
4772
5114
|
const usageObjects = objects.filter((object) => Object.keys(object).some((key) => key.toLowerCase().includes("token") || key.toLowerCase() === "usage"));
|
|
4773
5115
|
const candidates = usageObjects.length ? usageObjects : objects;
|
|
5116
|
+
const detailObjects = candidates.flatMap((candidate) => [firstObject6(candidate.input_tokens_details), firstObject6(candidate.inputTokensDetails), firstObject6(candidate.cache_write_tokens_details)].filter((item) => Boolean(item)));
|
|
5117
|
+
const cacheCreation30m = firstNumberByKeys(detailObjects, ["ttl_30m_tokens", "ttl30mTokens"]) ?? 0;
|
|
5118
|
+
const cacheCreation12h = firstNumberByKeys(detailObjects, ["ttl_12h_tokens", "ttl12hTokens"]) ?? 0;
|
|
5119
|
+
const cacheCreationDeclared = firstNumberByKeys(candidates, ["cacheCreationInputTokens", "cache_creation_input_tokens", "cacheWriteInputTokens", "cache_write_input_tokens", "cache_write_tokens"]) ?? firstNumberByKeys(detailObjects, ["cache_write_tokens", "cacheWriteTokens"]) ?? 0;
|
|
4774
5120
|
return {
|
|
4775
5121
|
inputTokens: firstNumberByKeys(candidates, ["inputTokens", "input_tokens", "promptTokens", "prompt_tokens", "totalUsedTokens"]) ?? 0,
|
|
4776
|
-
cachedInputTokens: firstNumberByKeys(candidates, ["cachedInputTokens", "cached_input_tokens", "cacheReadInputTokens", "cache_read_input_tokens"]) ?? 0,
|
|
5122
|
+
cachedInputTokens: firstNumberByKeys(candidates, ["cachedInputTokens", "cached_input_tokens", "cached_tokens", "cacheReadInputTokens", "cache_read_input_tokens"]) ?? firstNumberByKeys(detailObjects, ["cached_tokens", "cachedTokens"]) ?? 0,
|
|
5123
|
+
cacheCreationInputTokens: Math.max(cacheCreationDeclared, cacheCreation30m + cacheCreation12h),
|
|
5124
|
+
cacheCreationInputTokens30m: cacheCreation30m,
|
|
5125
|
+
cacheCreationInputTokens12h: cacheCreation12h,
|
|
4777
5126
|
outputTokens: firstNumberByKeys(candidates, ["outputTokens", "output_tokens", "completionTokens", "completion_tokens"]) ?? 0,
|
|
4778
5127
|
reasoningTokens: firstNumberByKeys(candidates, ["reasoningTokens", "reasoning_tokens", "reasoningOutputTokens", "reasoning_output_tokens"]) ?? 0,
|
|
4779
5128
|
totalTokens: firstNumberByKeys(candidates, ["totalTokens", "total_tokens", "tokens", "tokenCount", "token_count"]) ?? 0,
|
|
@@ -4829,6 +5178,9 @@ function candidateToUsage(candidate, index, options) {
|
|
|
4829
5178
|
provider: providerForModel2(breakdown.model ?? ""),
|
|
4830
5179
|
inputTokens: breakdown.inputTokens,
|
|
4831
5180
|
cachedInputTokens: breakdown.cachedInputTokens,
|
|
5181
|
+
cacheCreationInputTokens: breakdown.cacheCreationInputTokens || void 0,
|
|
5182
|
+
cacheCreationInputTokens30m: breakdown.cacheCreationInputTokens30m || void 0,
|
|
5183
|
+
cacheCreationInputTokens12h: breakdown.cacheCreationInputTokens12h || void 0,
|
|
4832
5184
|
outputTokens: breakdown.outputTokens,
|
|
4833
5185
|
reasoningTokens: breakdown.reasoningTokens,
|
|
4834
5186
|
reasoningOutputTokens: breakdown.reasoningTokens,
|
|
@@ -4875,12 +5227,13 @@ function candidateToUsage(candidate, index, options) {
|
|
|
4875
5227
|
}
|
|
4876
5228
|
}
|
|
4877
5229
|
};
|
|
4878
|
-
const
|
|
5230
|
+
const estimate = hasTokenBreakdown ? calculateRequestCosts([usage], options.pricing) : void 0;
|
|
5231
|
+
const pricedCost = estimate?.costUsd;
|
|
4879
5232
|
const estimatedCostUsd = pricedCost ?? breakdown.estimatedCostUsd;
|
|
4880
5233
|
return {
|
|
4881
5234
|
...usage,
|
|
4882
5235
|
estimatedCostUsd,
|
|
4883
|
-
warnings: estimatedCostUsd === void 0 ? [...usage.warnings, breakdown.hasEstimatedPromptTokens ? "cursor_cost_not_estimated_from_prompt_tokens" : hasTokenBreakdown ? "unknown_pricing" : "unknown_cost"] : usage.warnings
|
|
5236
|
+
warnings: estimatedCostUsd === void 0 ? [.../* @__PURE__ */ new Set([...usage.warnings, breakdown.hasEstimatedPromptTokens ? "cursor_cost_not_estimated_from_prompt_tokens" : hasTokenBreakdown ? "unknown_pricing" : "unknown_cost", ...estimate?.warnings ?? [], ...estimate?.assumptions ?? []])] : [.../* @__PURE__ */ new Set([...usage.warnings, ...estimate?.assumptions ?? []])]
|
|
4884
5237
|
};
|
|
4885
5238
|
}
|
|
4886
5239
|
function inferProjectPathFromCursorPath(filePath, cursorHome) {
|
|
@@ -4994,6 +5347,9 @@ function emptyBreakdown4() {
|
|
|
4994
5347
|
model: void 0,
|
|
4995
5348
|
inputTokens: 0,
|
|
4996
5349
|
cachedInputTokens: 0,
|
|
5350
|
+
cacheCreationInputTokens: 0,
|
|
5351
|
+
cacheCreationInputTokens30m: 0,
|
|
5352
|
+
cacheCreationInputTokens12h: 0,
|
|
4997
5353
|
outputTokens: 0,
|
|
4998
5354
|
reasoningTokens: 0,
|
|
4999
5355
|
totalTokens: 0,
|
|
@@ -5271,7 +5627,9 @@ function readPricingData() {
|
|
|
5271
5627
|
const pricingPath = resolvePricingPath(config);
|
|
5272
5628
|
return {
|
|
5273
5629
|
path: pricingPath,
|
|
5274
|
-
models: loadPricingTable(pricingPath)
|
|
5630
|
+
models: loadPricingTable(pricingPath),
|
|
5631
|
+
defaults: defaultPricing,
|
|
5632
|
+
contextRules: pricingContextRules()
|
|
5275
5633
|
};
|
|
5276
5634
|
}
|
|
5277
5635
|
function readConfigData() {
|
|
@@ -5295,7 +5653,9 @@ function writePricingData(models) {
|
|
|
5295
5653
|
clearScanCache();
|
|
5296
5654
|
return {
|
|
5297
5655
|
path: pricingPath,
|
|
5298
|
-
models: loadPricingTable(pricingPath)
|
|
5656
|
+
models: loadPricingTable(pricingPath),
|
|
5657
|
+
defaults: defaultPricing,
|
|
5658
|
+
contextRules: pricingContextRules()
|
|
5299
5659
|
};
|
|
5300
5660
|
}
|
|
5301
5661
|
function clearRepoSpendLocalData() {
|
|
@@ -6366,12 +6726,18 @@ function recentCommandLines(lines) {
|
|
|
6366
6726
|
function createServer(options = {}) {
|
|
6367
6727
|
const app = Fastify({ logger: false });
|
|
6368
6728
|
app.addHook("onRequest", async (request, reply) => {
|
|
6369
|
-
|
|
6370
|
-
if (
|
|
6371
|
-
|
|
6729
|
+
const path11 = guardRequestPath(request);
|
|
6730
|
+
if (isProtectedLocalMutation(request.method, path11)) {
|
|
6731
|
+
if (!isAllowedLocalHost(request.headers.host)) {
|
|
6732
|
+
return reply.code(403).send({ error: "Mutating RepoSpend local settings requires a localhost Host header." });
|
|
6733
|
+
}
|
|
6734
|
+
if (!isAllowedLocalOrigin(request.headers.origin)) {
|
|
6735
|
+
return reply.code(403).send({ error: "Mutating RepoSpend local settings requires a localhost Origin." });
|
|
6736
|
+
}
|
|
6737
|
+
return;
|
|
6372
6738
|
}
|
|
6373
|
-
if (!
|
|
6374
|
-
return reply.code(403).send({ error: "
|
|
6739
|
+
if (isSensitiveLocalRead(request.method, path11) && request.headers.host && !isAllowedLocalHost(request.headers.host)) {
|
|
6740
|
+
return reply.code(403).send({ error: "Reading RepoSpend local data requires a localhost Host header." });
|
|
6375
6741
|
}
|
|
6376
6742
|
});
|
|
6377
6743
|
app.get("/api/health", async () => ({ ok: true, app: "RepoSpend", version: readPackageVersion() }));
|
|
@@ -6486,7 +6852,26 @@ function isAddressInUse(error) {
|
|
|
6486
6852
|
}
|
|
6487
6853
|
function isProtectedLocalMutation(method, url) {
|
|
6488
6854
|
if (method !== "PUT" && method !== "POST" && method !== "DELETE") return false;
|
|
6489
|
-
|
|
6855
|
+
const pathname = routerNormalizedPath(url);
|
|
6856
|
+
return pathname.startsWith("/api/settings/") || pathname.startsWith("/api/cache/");
|
|
6857
|
+
}
|
|
6858
|
+
var SENSITIVE_LOCAL_READS = /* @__PURE__ */ new Set(["/api/dashboard", "/api/export.json", "/api/export.csv"]);
|
|
6859
|
+
function isSensitiveLocalRead(method, url) {
|
|
6860
|
+
if (method !== "GET" && method !== "HEAD") return false;
|
|
6861
|
+
return SENSITIVE_LOCAL_READS.has(routerNormalizedPath(url));
|
|
6862
|
+
}
|
|
6863
|
+
function guardRequestPath(request) {
|
|
6864
|
+
const matched = request.routeOptions?.url;
|
|
6865
|
+
return typeof matched === "string" ? matched : routerNormalizedPath(request.url);
|
|
6866
|
+
}
|
|
6867
|
+
function routerNormalizedPath(url) {
|
|
6868
|
+
const absoluteStripped = url.replace(/^[a-zA-Z][a-zA-Z0-9+.-]*:\/\/[^/?#]*/, "");
|
|
6869
|
+
const raw = absoluteStripped.split("?")[0] ?? "";
|
|
6870
|
+
try {
|
|
6871
|
+
return decodeURIComponent(raw);
|
|
6872
|
+
} catch {
|
|
6873
|
+
return raw;
|
|
6874
|
+
}
|
|
6490
6875
|
}
|
|
6491
6876
|
function isAllowedLocalHost(hostHeader) {
|
|
6492
6877
|
if (!hostHeader) return false;
|
|
@@ -6533,11 +6918,15 @@ function validatePricingBody(body) {
|
|
|
6533
6918
|
const cachedInputPerMillion = finiteNumber(row.cachedInputPerMillion);
|
|
6534
6919
|
const cacheCreationInput5mPerMillion = finiteNumber(row.cacheCreationInput5mPerMillion);
|
|
6535
6920
|
const cacheCreationInput1hPerMillion = finiteNumber(row.cacheCreationInput1hPerMillion);
|
|
6921
|
+
const cacheCreationInput30mPerMillion = finiteNumber(row.cacheCreationInput30mPerMillion);
|
|
6922
|
+
const cacheCreationInput12hPerMillion = finiteNumber(row.cacheCreationInput12hPerMillion);
|
|
6536
6923
|
const cacheCreationInputPerMillion = finiteNumber(row.cacheCreationInputPerMillion);
|
|
6537
6924
|
const reasoningOutputPerMillion = finiteNumber(row.reasoningOutputPerMillion);
|
|
6538
6925
|
if (cachedInputPerMillion !== void 0) modelPricing.cachedInputPerMillion = cachedInputPerMillion;
|
|
6539
6926
|
if (cacheCreationInput5mPerMillion !== void 0) modelPricing.cacheCreationInput5mPerMillion = cacheCreationInput5mPerMillion;
|
|
6540
6927
|
if (cacheCreationInput1hPerMillion !== void 0) modelPricing.cacheCreationInput1hPerMillion = cacheCreationInput1hPerMillion;
|
|
6928
|
+
if (cacheCreationInput30mPerMillion !== void 0) modelPricing.cacheCreationInput30mPerMillion = cacheCreationInput30mPerMillion;
|
|
6929
|
+
if (cacheCreationInput12hPerMillion !== void 0) modelPricing.cacheCreationInput12hPerMillion = cacheCreationInput12hPerMillion;
|
|
6541
6930
|
if (cacheCreationInputPerMillion !== void 0) modelPricing.cacheCreationInputPerMillion = cacheCreationInputPerMillion;
|
|
6542
6931
|
if (reasoningOutputPerMillion !== void 0) modelPricing.reasoningOutputPerMillion = reasoningOutputPerMillion;
|
|
6543
6932
|
if (typeof row.note === "string") modelPricing.note = row.note;
|