@aiusage/cli 1.8.2 → 1.8.4
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/dist/cli.js +119 -12
- package/package.json +1 -1
- package/pricing-catalog.json +144 -1
package/dist/cli.js
CHANGED
|
@@ -413,6 +413,39 @@ import { createInterface as createInterface2 } from "node:readline";
|
|
|
413
413
|
var anthropic = {
|
|
414
414
|
"claude-code": {
|
|
415
415
|
models: {
|
|
416
|
+
// https://platform.claude.com/docs/en/about-claude/pricing (2026-11-14)
|
|
417
|
+
"claude-fable-5-2": {
|
|
418
|
+
currency: "USD",
|
|
419
|
+
effective_from: "2026-11-09",
|
|
420
|
+
input_per_million: 8,
|
|
421
|
+
output_per_million: 40,
|
|
422
|
+
cache_write_5m_per_million: 10,
|
|
423
|
+
cache_write_1h_per_million: 16,
|
|
424
|
+
cached_input_per_million: 0.2
|
|
425
|
+
},
|
|
426
|
+
// https://platform.claude.com/docs/en/models/haiku-5-5/overview (2026-11-14)
|
|
427
|
+
"claude-haiku-5-5": {
|
|
428
|
+
currency: "USD",
|
|
429
|
+
effective_from: "2026-10-07",
|
|
430
|
+
input_per_million: 0.1,
|
|
431
|
+
output_per_million: 0.5,
|
|
432
|
+
cache_write_5m_per_million: 0.125,
|
|
433
|
+
cache_write_1h_per_million: 0.2,
|
|
434
|
+
cached_input_per_million: 0.01,
|
|
435
|
+
tiers: [
|
|
436
|
+
{ threshold: 1e5, input_per_million: 0.1, output_per_million: 0.5, cached_input_per_million: 0.01, cache_write_5m_per_million: 0.125, cache_write_1h_per_million: 0.2 },
|
|
437
|
+
{ input_per_million: 0.5, output_per_million: 2.5, cached_input_per_million: 0.05, cache_write_5m_per_million: 0.625, cache_write_1h_per_million: 1 }
|
|
438
|
+
]
|
|
439
|
+
},
|
|
440
|
+
// https://platform.claude.com/docs/en/about-claude/pricing (2026-10-20)
|
|
441
|
+
"claude-sonnet-5-5": {
|
|
442
|
+
currency: "USD",
|
|
443
|
+
input_per_million: 2,
|
|
444
|
+
output_per_million: 10,
|
|
445
|
+
cache_write_5m_per_million: 2.5,
|
|
446
|
+
cache_write_1h_per_million: 4,
|
|
447
|
+
cached_input_per_million: 0.1
|
|
448
|
+
},
|
|
416
449
|
"claude-fable-5-1": {
|
|
417
450
|
currency: "USD",
|
|
418
451
|
effective_from: "2026-09-01",
|
|
@@ -566,6 +599,60 @@ var openai = {
|
|
|
566
599
|
codex: {
|
|
567
600
|
models: {
|
|
568
601
|
// ── GPT-6 系列 ──
|
|
602
|
+
// https://developers.openai.com/api/docs/models/gpt-6.2-astra (2026-11-18)
|
|
603
|
+
"gpt-6.2-astra": {
|
|
604
|
+
currency: "USD",
|
|
605
|
+
effective_from: "2026-11-17",
|
|
606
|
+
notes: "prompts over 272K input tokens bill the full request at long-context rates; cache writes use the 30-minute rate",
|
|
607
|
+
input_per_million: 8,
|
|
608
|
+
cached_input_per_million: 0.4,
|
|
609
|
+
cache_write_per_million: 10,
|
|
610
|
+
output_per_million: 40,
|
|
611
|
+
tiers: [
|
|
612
|
+
{ threshold: 272e3, input_per_million: 8, cached_input_per_million: 0.4, cache_write_per_million: 10, output_per_million: 40 },
|
|
613
|
+
{ input_per_million: 16, cached_input_per_million: 0.8, cache_write_per_million: 20, output_per_million: 60 }
|
|
614
|
+
]
|
|
615
|
+
},
|
|
616
|
+
// https://developers.openai.com/api/docs/models/gpt-6.2-sol (2026-11-18)
|
|
617
|
+
"gpt-6.2-sol": {
|
|
618
|
+
currency: "USD",
|
|
619
|
+
effective_from: "2026-11-17",
|
|
620
|
+
notes: "prompts over 272K input tokens bill the full request at long-context rates; cache writes use the 30-minute rate",
|
|
621
|
+
input_per_million: 2,
|
|
622
|
+
cached_input_per_million: 0.1,
|
|
623
|
+
cache_write_per_million: 2.5,
|
|
624
|
+
output_per_million: 10,
|
|
625
|
+
tiers: [
|
|
626
|
+
{ threshold: 272e3, input_per_million: 2, cached_input_per_million: 0.1, cache_write_per_million: 2.5, output_per_million: 10 },
|
|
627
|
+
{ input_per_million: 4, cached_input_per_million: 0.2, cache_write_per_million: 5, output_per_million: 15 }
|
|
628
|
+
]
|
|
629
|
+
},
|
|
630
|
+
// https://developers.openai.com/api/docs/models/gpt-6.1-terra (2026-11-14)
|
|
631
|
+
"gpt-6.1-terra": {
|
|
632
|
+
currency: "USD",
|
|
633
|
+
notes: "prompts over 272K input tokens bill the full request at long-context rates; cache write uses the 30-minute rate",
|
|
634
|
+
input_per_million: 1,
|
|
635
|
+
cached_input_per_million: 0.05,
|
|
636
|
+
cache_write_per_million: 1.25,
|
|
637
|
+
output_per_million: 5,
|
|
638
|
+
tiers: [
|
|
639
|
+
{ threshold: 272e3, input_per_million: 1, cached_input_per_million: 0.05, cache_write_per_million: 1.25, output_per_million: 5 },
|
|
640
|
+
{ input_per_million: 2, cached_input_per_million: 0.1, cache_write_per_million: 2.5, output_per_million: 7.5 }
|
|
641
|
+
]
|
|
642
|
+
},
|
|
643
|
+
// https://developers.openai.com/api/docs/models/gpt-6.1-sol (2026-10-20)
|
|
644
|
+
"gpt-6.1-sol": {
|
|
645
|
+
currency: "USD",
|
|
646
|
+
notes: "prompts over 272K input tokens bill the full request at long-context rates",
|
|
647
|
+
input_per_million: 2,
|
|
648
|
+
cached_input_per_million: 0.1,
|
|
649
|
+
cache_write_per_million: 2.5,
|
|
650
|
+
output_per_million: 10,
|
|
651
|
+
tiers: [
|
|
652
|
+
{ threshold: 272e3, input_per_million: 2, cached_input_per_million: 0.1, cache_write_per_million: 2.5, output_per_million: 10 },
|
|
653
|
+
{ input_per_million: 4, cached_input_per_million: 0.2, cache_write_per_million: 5, output_per_million: 15 }
|
|
654
|
+
]
|
|
655
|
+
},
|
|
569
656
|
"gpt-6-astra": {
|
|
570
657
|
currency: "USD",
|
|
571
658
|
notes: "prompts over 272K input tokens bill the full request at long-context rates",
|
|
@@ -949,6 +1036,13 @@ var google = {
|
|
|
949
1036
|
"gemini-cli": {
|
|
950
1037
|
models: {
|
|
951
1038
|
// ── Gemini 3.x 系列 ──
|
|
1039
|
+
// https://aistudio.google.com/docs/pricing#gemini-3.5-flash-lite (2026-11-14)
|
|
1040
|
+
"gemini-3.5-flash-lite": {
|
|
1041
|
+
currency: "USD",
|
|
1042
|
+
input_per_million: 0.3,
|
|
1043
|
+
output_per_million: 2.5,
|
|
1044
|
+
cached_input_per_million: 0.03
|
|
1045
|
+
},
|
|
952
1046
|
"gemini-3.5-flash": {
|
|
953
1047
|
currency: "USD",
|
|
954
1048
|
input_per_million: 1.5,
|
|
@@ -1530,7 +1624,7 @@ var inflection = {
|
|
|
1530
1624
|
};
|
|
1531
1625
|
|
|
1532
1626
|
// ../shared/src/pricing/catalog.ts
|
|
1533
|
-
var PRICING_VERSION = "2026-
|
|
1627
|
+
var PRICING_VERSION = "2026-11-18-gpt-6-2-v1", aliases = {
|
|
1534
1628
|
"claude-fibre-5": "claude-fable-5",
|
|
1535
1629
|
"claude-fiber-5": "claude-fable-5",
|
|
1536
1630
|
"claude-ops-5": "claude-opus-5",
|
|
@@ -1571,12 +1665,17 @@ var PRICING_VERSION = "2026-09-24-gpt-6-opus-5-5-grok-4-7-v1", aliases = {
|
|
|
1571
1665
|
|
|
1572
1666
|
// ../shared/src/pricing/calculate.ts
|
|
1573
1667
|
var FAST_MULTIPLIERS = {
|
|
1668
|
+
"claude-fable-5-2": 2,
|
|
1574
1669
|
"claude-opus-5-5": 2,
|
|
1575
1670
|
"claude-opus-5": 2,
|
|
1576
1671
|
"claude-opus-4-8": 2,
|
|
1577
1672
|
"claude-opus-4-7": 6,
|
|
1578
1673
|
"grok-4.7": 2
|
|
1579
|
-
}, OPENAI_CODEX_TIER_MULTIPLIERS = {
|
|
1674
|
+
}, GPT_6_2_ULTRAFAST_PROMOTION_END = Date.parse("2027-03-01T00:00:00-08:00"), OPENAI_CODEX_TIER_MULTIPLIERS = {
|
|
1675
|
+
"gpt-6.2-astra": 2,
|
|
1676
|
+
"gpt-6.2-sol": 2,
|
|
1677
|
+
"gpt-6.1-sol": 2,
|
|
1678
|
+
"gpt-6.1-terra": 2,
|
|
1580
1679
|
"gpt-6-astra": 2,
|
|
1581
1680
|
"gpt-6-sol": 2,
|
|
1582
1681
|
"gpt-6-luna": 2,
|
|
@@ -1587,10 +1686,11 @@ var FAST_MULTIPLIERS = {
|
|
|
1587
1686
|
"gpt-5.4": 2
|
|
1588
1687
|
};
|
|
1589
1688
|
function splitServiceTierSuffix(model) {
|
|
1590
|
-
|
|
1689
|
+
let gpt62 = /^(gpt-6\.2-(?:astra|sol))-(ultrafast|batch)$/.exec(model);
|
|
1690
|
+
return gpt62 ? { baseModel: gpt62[1], tier: gpt62[2] === "ultrafast" ? "ultrafast" : "batch" } : model === "gpt-6.1-sol-ultrafast" ? { baseModel: model.replace(/-ultrafast$/, ""), tier: "ultrafast" } : model.endsWith("-priority") ? { baseModel: model.replace(/-priority$/, ""), tier: "priority" } : model.endsWith("-fast") ? { baseModel: model.replace(/-fast$/, ""), tier: "fast" } : { baseModel: model, tier: null };
|
|
1591
1691
|
}
|
|
1592
1692
|
function getServiceTierMultiplier(provider, product, resolvedModel, tier) {
|
|
1593
|
-
return tier ? provider === "openai" && product === "codex" ? tier === "fast" && resolvedModel.startsWith("gpt-5.6-") ? 1 : OPENAI_CODEX_TIER_MULTIPLIERS[resolvedModel] ?? 1 : tier === "fast" ? FAST_MULTIPLIERS[resolvedModel] ?? 1 : 1 : 1;
|
|
1693
|
+
return tier ? provider === "openai" && product === "codex" ? tier === "ultrafast" ? 6 : tier === "batch" ? 0.5 : tier === "fast" && resolvedModel.startsWith("gpt-5.6-") ? 1 : OPENAI_CODEX_TIER_MULTIPLIERS[resolvedModel] ?? 1 : tier === "fast" ? FAST_MULTIPLIERS[resolvedModel] ?? 1 : 1 : 1;
|
|
1594
1694
|
}
|
|
1595
1695
|
function resolveModelPricing(catalog2, provider, product, model) {
|
|
1596
1696
|
let models2 = catalog2.providers[provider]?.[product]?.models;
|
|
@@ -1643,7 +1743,12 @@ function calculateCost(provider, product, model, tokens, options = {}) {
|
|
|
1643
1743
|
};
|
|
1644
1744
|
let cw5Rate = unit.cache_write_5m_per_million ?? pricing.cache_write_5m_per_million ?? 0, cw1hRate = unit.cache_write_1h_per_million ?? pricing.cache_write_1h_per_million ?? 0, hasGenericCacheWriteRate = unit.cache_write_per_million !== void 0 || pricing.cache_write_per_million !== void 0, genericCwRate = unit.cache_write_per_million ?? pricing.cache_write_per_million ?? 0, cachedRate = unit.cached_input_per_million ?? pricing.cached_input_per_million ?? 0, cacheWriteCost = hasGenericCacheWriteRate ? tokens.cacheWriteTokens / 1e6 * genericCwRate : (tokens.cacheWrite5mTokens ?? tokens.cacheWriteTokens) / 1e6 * cw5Rate + (tokens.cacheWrite1hTokens ?? 0) / 1e6 * cw1hRate, raw = tokens.inputTokens / 1e6 * (unit.input_per_million ?? 0) + tokens.cachedInputTokens / 1e6 * (cachedRate ?? 0) + cacheWriteCost + tokens.outputTokens / 1e6 * (unit.output_per_million ?? 0);
|
|
1645
1745
|
raw = toUsd(raw, pricing.currency, cat);
|
|
1646
|
-
let
|
|
1746
|
+
let multiplier = getServiceTierMultiplier(provider, product, resolvedModel, tier);
|
|
1747
|
+
if (provider === "openai" && product === "codex" && tier === "ultrafast" && (resolvedModel === "gpt-6.2-astra" || resolvedModel === "gpt-6.2-sol")) {
|
|
1748
|
+
let { usageTime } = options;
|
|
1749
|
+
typeof usageTime == "string" ? (multiplier = usageTime < "2027-03-01" ? 3 : 6, (usageTime === "2027-02-28" || usageTime === "2027-03-01") && (costStatus = "estimated")) : (multiplier = (usageTime ?? Date.now()) < GPT_6_2_ULTRAFAST_PROMOTION_END ? 3 : 6, usageTime === void 0 && (costStatus = "estimated"));
|
|
1750
|
+
}
|
|
1751
|
+
let finalCost = raw * multiplier;
|
|
1647
1752
|
return {
|
|
1648
1753
|
estimatedCostUsd: Math.round(finalCost * 1e4) / 1e4,
|
|
1649
1754
|
costStatus,
|
|
@@ -1745,7 +1850,7 @@ async function processCodexFile(filePath, projectAliases, serviceTier) {
|
|
|
1745
1850
|
// Reasoning is billed at the output rate, so pricing still uses the
|
|
1746
1851
|
// inclusive raw Codex output total.
|
|
1747
1852
|
outputTokens: tokens.output
|
|
1748
|
-
}), exactEventCost = eventCost.costStatus === "exact" ? eventCost.estimatedCostUsd : void 0;
|
|
1853
|
+
}, { usageTime: ts.getTime() }), exactEventCost = eventCost.costStatus === "exact" ? eventCost.estimatedCostUsd : void 0;
|
|
1749
1854
|
events.push({
|
|
1750
1855
|
usageDate,
|
|
1751
1856
|
signature,
|
|
@@ -4146,7 +4251,7 @@ async function buildLocalReport(range, options = {}) {
|
|
|
4146
4251
|
if (hasData) {
|
|
4147
4252
|
daysWithData += 1;
|
|
4148
4253
|
for (let breakdown of result.breakdowns) {
|
|
4149
|
-
let breakdownTotals = toBreakdownTotals(breakdown, pricingWarnings, options.pricingCatalog);
|
|
4254
|
+
let breakdownTotals = toBreakdownTotals(breakdown, pricingWarnings, options.pricingCatalog, usageDate);
|
|
4150
4255
|
dayTotals.estimatedCostUsd += breakdownTotals.estimatedCostUsd, mergeTotals(totals, breakdownTotals), mergeTotals(getOrCreate(bySource, `${breakdown.provider}/${breakdown.product}`), breakdownTotals), mergeTotals(
|
|
4151
4256
|
getOrCreate(byModel, `${breakdown.provider}/${breakdown.product}|${breakdown.model}`),
|
|
4152
4257
|
breakdownTotals
|
|
@@ -4559,8 +4664,8 @@ function createEmptyTotals2() {
|
|
|
4559
4664
|
function mergeTotals(target, source) {
|
|
4560
4665
|
return target.eventCount += source.eventCount, target.inputTokens += source.inputTokens, target.cachedInputTokens += source.cachedInputTokens, target.cacheWriteTokens += source.cacheWriteTokens, target.outputTokens += source.outputTokens, target.reasoningOutputTokens += source.reasoningOutputTokens, target.totalTokens += source.totalTokens, target.estimatedCostUsd += source.estimatedCostUsd, target;
|
|
4561
4666
|
}
|
|
4562
|
-
function toBreakdownTotals(breakdown, warnings, pricingCatalog) {
|
|
4563
|
-
let estimatedCostUsd = calculateBreakdownCost(breakdown, warnings, pricingCatalog);
|
|
4667
|
+
function toBreakdownTotals(breakdown, warnings, pricingCatalog, usageDate) {
|
|
4668
|
+
let estimatedCostUsd = calculateBreakdownCost(breakdown, warnings, pricingCatalog, usageDate);
|
|
4564
4669
|
return {
|
|
4565
4670
|
eventCount: breakdown.eventCount,
|
|
4566
4671
|
inputTokens: breakdown.inputTokens,
|
|
@@ -4572,7 +4677,7 @@ function toBreakdownTotals(breakdown, warnings, pricingCatalog) {
|
|
|
4572
4677
|
estimatedCostUsd
|
|
4573
4678
|
};
|
|
4574
4679
|
}
|
|
4575
|
-
function calculateBreakdownCost(breakdown, warnings, pricingCatalog) {
|
|
4680
|
+
function calculateBreakdownCost(breakdown, warnings, pricingCatalog, usageDate) {
|
|
4576
4681
|
let effectivePricingVersion = pricingCatalog?.version ?? PRICING_VERSION, sourceCostMatchesCatalog = breakdown.product === "trae-intl" || breakdown.product === "opencode" || breakdown.pricingVersion == null || breakdown.pricingVersion === effectivePricingVersion;
|
|
4577
4682
|
if (breakdown.costUSD != null && breakdown.costUSD > 0 && sourceCostMatchesCatalog)
|
|
4578
4683
|
return breakdown.costUSD;
|
|
@@ -4586,11 +4691,13 @@ function calculateBreakdownCost(breakdown, warnings, pricingCatalog) {
|
|
|
4586
4691
|
cacheWriteTokens: breakdown.cacheWriteTokens,
|
|
4587
4692
|
cacheWrite5mTokens: breakdown.cacheWrite5mTokens,
|
|
4588
4693
|
cacheWrite1hTokens: breakdown.cacheWrite1hTokens,
|
|
4589
|
-
|
|
4694
|
+
// Codex stores reasoning separately from visible output in the breakdown.
|
|
4695
|
+
outputTokens: breakdown.outputTokens + (breakdown.product === "codex" ? breakdown.reasoningOutputTokens ?? 0 : 0)
|
|
4590
4696
|
},
|
|
4591
4697
|
{
|
|
4592
4698
|
...pricingCatalog ? { catalog: pricingCatalog } : {},
|
|
4593
|
-
requestCount: breakdown.eventCount
|
|
4699
|
+
requestCount: breakdown.eventCount,
|
|
4700
|
+
usageTime: usageDate
|
|
4594
4701
|
}
|
|
4595
4702
|
);
|
|
4596
4703
|
return result.costStatus === "unavailable" ? warnings.add(`${breakdown.provider}/${breakdown.product}/${breakdown.model} \u6682\u65E0\u5B9A\u4EF7\u914D\u7F6E\uFF0C\u5DF2\u8DF3\u8FC7\u6210\u672C\u4F30\u7B97\u3002`) : result.costStatus === "estimated" && result.resolvedModel && result.resolvedModel !== breakdown.model ? warnings.add(`${breakdown.model} \u5DF2\u6309 ${result.resolvedModel} \u7684\u516C\u5F00\u5355\u4EF7\u4F30\u7B97\u3002`) : result.costStatus === "estimated" && result.matchedTierIndex !== void 0 && warnings.add(`${breakdown.model} \u7684\u9636\u68AF\u4EF7\u683C\u5DF2\u6309\u6BCF\u4E8B\u4EF6\u5E73\u5747\u8F93\u5165\u91CF\u4F30\u7B97\u3002`), result.estimatedCostUsd;
|
package/package.json
CHANGED
package/pricing-catalog.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
{
|
|
2
|
-
"version": "2026-
|
|
2
|
+
"version": "2026-11-18-gpt-6-2-v1",
|
|
3
3
|
"fx": {
|
|
4
4
|
"CNY": 7.2
|
|
5
5
|
},
|
|
@@ -22,6 +22,49 @@
|
|
|
22
22
|
"anthropic": {
|
|
23
23
|
"claude-code": {
|
|
24
24
|
"models": {
|
|
25
|
+
"claude-fable-5-2": {
|
|
26
|
+
"currency": "USD",
|
|
27
|
+
"effective_from": "2026-11-09",
|
|
28
|
+
"input_per_million": 8,
|
|
29
|
+
"output_per_million": 40,
|
|
30
|
+
"cache_write_5m_per_million": 10,
|
|
31
|
+
"cache_write_1h_per_million": 16,
|
|
32
|
+
"cached_input_per_million": 0.2
|
|
33
|
+
},
|
|
34
|
+
"claude-haiku-5-5": {
|
|
35
|
+
"currency": "USD",
|
|
36
|
+
"effective_from": "2026-10-07",
|
|
37
|
+
"input_per_million": 0.1,
|
|
38
|
+
"output_per_million": 0.5,
|
|
39
|
+
"cache_write_5m_per_million": 0.125,
|
|
40
|
+
"cache_write_1h_per_million": 0.2,
|
|
41
|
+
"cached_input_per_million": 0.01,
|
|
42
|
+
"tiers": [
|
|
43
|
+
{
|
|
44
|
+
"threshold": 100000,
|
|
45
|
+
"input_per_million": 0.1,
|
|
46
|
+
"output_per_million": 0.5,
|
|
47
|
+
"cached_input_per_million": 0.01,
|
|
48
|
+
"cache_write_5m_per_million": 0.125,
|
|
49
|
+
"cache_write_1h_per_million": 0.2
|
|
50
|
+
},
|
|
51
|
+
{
|
|
52
|
+
"input_per_million": 0.5,
|
|
53
|
+
"output_per_million": 2.5,
|
|
54
|
+
"cached_input_per_million": 0.05,
|
|
55
|
+
"cache_write_5m_per_million": 0.625,
|
|
56
|
+
"cache_write_1h_per_million": 1
|
|
57
|
+
}
|
|
58
|
+
]
|
|
59
|
+
},
|
|
60
|
+
"claude-sonnet-5-5": {
|
|
61
|
+
"currency": "USD",
|
|
62
|
+
"input_per_million": 2,
|
|
63
|
+
"output_per_million": 10,
|
|
64
|
+
"cache_write_5m_per_million": 2.5,
|
|
65
|
+
"cache_write_1h_per_million": 4,
|
|
66
|
+
"cached_input_per_million": 0.1
|
|
67
|
+
},
|
|
25
68
|
"claude-fable-5-1": {
|
|
26
69
|
"currency": "USD",
|
|
27
70
|
"effective_from": "2026-09-01",
|
|
@@ -172,6 +215,100 @@
|
|
|
172
215
|
"openai": {
|
|
173
216
|
"codex": {
|
|
174
217
|
"models": {
|
|
218
|
+
"gpt-6.2-astra": {
|
|
219
|
+
"currency": "USD",
|
|
220
|
+
"effective_from": "2026-11-17",
|
|
221
|
+
"notes": "prompts over 272K input tokens bill the full request at long-context rates; cache writes use the 30-minute rate",
|
|
222
|
+
"input_per_million": 8,
|
|
223
|
+
"cached_input_per_million": 0.4,
|
|
224
|
+
"cache_write_per_million": 10,
|
|
225
|
+
"output_per_million": 40,
|
|
226
|
+
"tiers": [
|
|
227
|
+
{
|
|
228
|
+
"threshold": 272000,
|
|
229
|
+
"input_per_million": 8,
|
|
230
|
+
"cached_input_per_million": 0.4,
|
|
231
|
+
"cache_write_per_million": 10,
|
|
232
|
+
"output_per_million": 40
|
|
233
|
+
},
|
|
234
|
+
{
|
|
235
|
+
"input_per_million": 16,
|
|
236
|
+
"cached_input_per_million": 0.8,
|
|
237
|
+
"cache_write_per_million": 20,
|
|
238
|
+
"output_per_million": 60
|
|
239
|
+
}
|
|
240
|
+
]
|
|
241
|
+
},
|
|
242
|
+
"gpt-6.2-sol": {
|
|
243
|
+
"currency": "USD",
|
|
244
|
+
"effective_from": "2026-11-17",
|
|
245
|
+
"notes": "prompts over 272K input tokens bill the full request at long-context rates; cache writes use the 30-minute rate",
|
|
246
|
+
"input_per_million": 2,
|
|
247
|
+
"cached_input_per_million": 0.1,
|
|
248
|
+
"cache_write_per_million": 2.5,
|
|
249
|
+
"output_per_million": 10,
|
|
250
|
+
"tiers": [
|
|
251
|
+
{
|
|
252
|
+
"threshold": 272000,
|
|
253
|
+
"input_per_million": 2,
|
|
254
|
+
"cached_input_per_million": 0.1,
|
|
255
|
+
"cache_write_per_million": 2.5,
|
|
256
|
+
"output_per_million": 10
|
|
257
|
+
},
|
|
258
|
+
{
|
|
259
|
+
"input_per_million": 4,
|
|
260
|
+
"cached_input_per_million": 0.2,
|
|
261
|
+
"cache_write_per_million": 5,
|
|
262
|
+
"output_per_million": 15
|
|
263
|
+
}
|
|
264
|
+
]
|
|
265
|
+
},
|
|
266
|
+
"gpt-6.1-terra": {
|
|
267
|
+
"currency": "USD",
|
|
268
|
+
"notes": "prompts over 272K input tokens bill the full request at long-context rates; cache write uses the 30-minute rate",
|
|
269
|
+
"input_per_million": 1,
|
|
270
|
+
"cached_input_per_million": 0.05,
|
|
271
|
+
"cache_write_per_million": 1.25,
|
|
272
|
+
"output_per_million": 5,
|
|
273
|
+
"tiers": [
|
|
274
|
+
{
|
|
275
|
+
"threshold": 272000,
|
|
276
|
+
"input_per_million": 1,
|
|
277
|
+
"cached_input_per_million": 0.05,
|
|
278
|
+
"cache_write_per_million": 1.25,
|
|
279
|
+
"output_per_million": 5
|
|
280
|
+
},
|
|
281
|
+
{
|
|
282
|
+
"input_per_million": 2,
|
|
283
|
+
"cached_input_per_million": 0.1,
|
|
284
|
+
"cache_write_per_million": 2.5,
|
|
285
|
+
"output_per_million": 7.5
|
|
286
|
+
}
|
|
287
|
+
]
|
|
288
|
+
},
|
|
289
|
+
"gpt-6.1-sol": {
|
|
290
|
+
"currency": "USD",
|
|
291
|
+
"notes": "prompts over 272K input tokens bill the full request at long-context rates",
|
|
292
|
+
"input_per_million": 2,
|
|
293
|
+
"cached_input_per_million": 0.1,
|
|
294
|
+
"cache_write_per_million": 2.5,
|
|
295
|
+
"output_per_million": 10,
|
|
296
|
+
"tiers": [
|
|
297
|
+
{
|
|
298
|
+
"threshold": 272000,
|
|
299
|
+
"input_per_million": 2,
|
|
300
|
+
"cached_input_per_million": 0.1,
|
|
301
|
+
"cache_write_per_million": 2.5,
|
|
302
|
+
"output_per_million": 10
|
|
303
|
+
},
|
|
304
|
+
{
|
|
305
|
+
"input_per_million": 4,
|
|
306
|
+
"cached_input_per_million": 0.2,
|
|
307
|
+
"cache_write_per_million": 5,
|
|
308
|
+
"output_per_million": 15
|
|
309
|
+
}
|
|
310
|
+
]
|
|
311
|
+
},
|
|
175
312
|
"gpt-6-astra": {
|
|
176
313
|
"currency": "USD",
|
|
177
314
|
"notes": "prompts over 272K input tokens bill the full request at long-context rates",
|
|
@@ -708,6 +845,12 @@
|
|
|
708
845
|
},
|
|
709
846
|
"gemini-cli": {
|
|
710
847
|
"models": {
|
|
848
|
+
"gemini-3.5-flash-lite": {
|
|
849
|
+
"currency": "USD",
|
|
850
|
+
"input_per_million": 0.3,
|
|
851
|
+
"output_per_million": 2.5,
|
|
852
|
+
"cached_input_per_million": 0.03
|
|
853
|
+
},
|
|
711
854
|
"gemini-3.5-flash": {
|
|
712
855
|
"currency": "USD",
|
|
713
856
|
"input_per_million": 1.5,
|