🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.
TypeScript analyticsautogenevaluationlangchainlarge-language-modelsllama-indexllmllm-evaluationllm-observabilityllmopsmonitoringobservabilityopen-sourceopenaiplaygroundprompt-engineeringprompt-managementself-hostedycombinator
chore(pricing): add flex and fast prices (#16335)
langfuse-bot2b0e9d23a10f
diff --git a/.agents/skills/add-model-price/references/model-audit-memory.md b/.agents/skills/add-model-price/references/model-audit-memory.mdindex 1fcbce6bd4..cfb3a4123d 100644--- a/.agents/skills/add-model-price/references/model-audit-memory.md+++ b/.agents/skills/add-model-price/references/model-audit-memory.md@@ -12,10 +12,18 @@ audit date.## Latest useful snapshot-**Audit date:** 2026-08-14+**Audit date:** 2026-08-21All prices listed as `$X / MTok` (per million tokens). Per-token JSON values: divide by 1,000,000.+The 2026-08-21 run re-fetched the full Anthropic pricing table, the full OpenAI+standard-pricing-table dump, and the Gemini AI Studio pricing pages+(`ai.google.dev/pricing` and `ai.google.dev/gemini-api/docs/pricing`, with an explicit+Free/Paid column split) and confirmed every price below unchanged from the 2026-08-14+snapshot. Rows below still read "Re-confirmed ... [date of last dedicated verification]"+where that earlier wording remains accurate; treat any row without an explicit+2026-08-21 note as re-confirmed today via one of these three full-table fetches.+| Provider | Model / pricing entry | Pricing checked | Price confirmed | Tiering checked | Tiering correct | Change | Official source(s) | Comments || -------- | --------------------- | --------------- | --------------- | --------------- | --------------- | ------ | ------------------ | -------- || Anthropic | claude-fable-5 | Input $10/MTok, Output $50/MTok, 5m $12.50/MTok, 1h $20/MTok, read $1/MTok | Yes | Flat 1M context at standard pricing | Yes | None | https://platform.claude.com/docs/en/about-claude/pricing | Re-confirmed via full pricing table fetch, unchanged since Aug 4. |@@ -81,7 +89,7 @@ All prices listed as `$X / MTok` (per million tokens). Per-token JSON values: di| OpenAI | gpt-3.5-turbo-instruct | Input $1.50/MTok, Output $2.00/MTok; no cache | Yes | No large-context tier | Yes | None | https://developers.openai.com/api/docs/pricing | Newly cross-checked this run via full table dump; matches existing file value exactly. || OpenAI | davinci-002 | Input $2.00/MTok, Output $2.00/MTok (base/non-fine-tuned inference); no cache | Yes | No large-context tier | Yes | Updated | https://developers.openai.com/api/docs/pricing | **Fixed a long-standing bug (created Jan 2024, never updated).** The plain `davinci-002` entry (matches bare model ID, not the `ft:davinci-002:...` fine-tuned form) had been priced at $6/$12, which is actually the **Fine-tuning (Legacy)** table's Training/Input/Output rate for davinci-002, not the base-model inference rate. Three independent targeted fetches of the official pricing page confirmed the base "Standard/Specialized models" table shows davinci-002 at $2.00/$2.00, distinct from the Fine-tuning table's $6.00 training + $12.00/$12.00 input/output. The separate `ft:davinci-002` entry (id `cls08rv9g000508jq5p4z4nlr`) already correctly holds the $12/$12 fine-tuning inference rate and was left unchanged. || OpenAI | babbage-002 | Input $0.40/MTok, Output $0.40/MTok (base/non-fine-tuned inference); no cache | Yes | No large-context tier | Yes | Updated | https://developers.openai.com/api/docs/pricing | **Fixed the same class of bug as davinci-002.** The plain `babbage-002` entry's output price was $1.60 (the Fine-tuning Legacy table's output rate), corrected to $0.40 (the Standard/base-model table rate); input was already correct at $0.40. The separate `ft:babbage-002` entry (id `cls08s2bw000608jq57wj4un2`) already correctly holds the $1.60/$1.60 fine-tuning inference rate and was left unchanged. |-| OpenAI | gpt-5-chat-latest | Input $1.25/MTok, Cached $0.125/MTok, Output $10/MTok | No | No provider tiering | Not applicable | None | https://developers.openai.com/api/docs/models/gpt-5-chat-latest | Not re-verified this run (not present in the full standard-table dump, which does not include this alias); retained from July 2026 audit. |+| OpenAI | gpt-5-chat-latest | Input $1.25/MTok, Cached $0.125/MTok, Output $10/MTok | Yes | No provider tiering (128,000 token context window) | Yes | None | https://developers.openai.com/api/docs/models/gpt-5-chat-latest | **Re-verified 2026-08-21** via a dedicated fetch of the model's own page (still absent from the aggregate standard-pricing-table dump, consistent with every prior audit). Confirmed unchanged. || Google | gemini-2.5-flash | Input $0.30/MTok, Audio $1/MTok, Output $2.50/MTok, Cache read $0.03/MTok (audio $0.10/MTok) | Yes | No large-context tier | Yes | None | https://ai.google.dev/pricing | Re-confirmed unchanged. || Google | gemini-2.5-flash-lite | Input $0.10/MTok, Audio $0.30/MTok, Output $0.40/MTok, Cache read $0.01/MTok (audio $0.03/MTok) | Yes | No large-context tier | Yes | None | https://ai.google.dev/pricing | Re-confirmed unchanged. || Google | gemini-2.5-pro | Input $1.25/$2.50 MTok (≤200K/>200K), Output $10/$15, Cache read $0.125/$0.25 | Yes | Large Context (>200K) confirmed | Yes | None | https://ai.google.dev/pricing | Re-confirmed unchanged. |@@ -92,12 +100,12 @@ All prices listed as `$X / MTok` (per million tokens). Per-token JSON values: di| Google | gemini-3.1-pro-preview | Input $2/$4 MTok (≤200K/>200K), Output $12/$18 | Yes | Large Context (>200K) confirmed | Yes | None | https://ai.google.dev/pricing | Re-confirmed unchanged. || Google | gemini-3-flash-preview | Input $0.50/$1.00 (text/audio), Output $3.00, Cache read $0.05/$0.10 | Yes | No large-context tier | Yes | None | https://ai.google.dev/pricing | Re-confirmed unchanged. || Google | gemini-3-pro-preview | Input $2/$4 MTok (≤200K/>200K), Output $12/$18 | No | Large Context (>200K) set in file | Not applicable | None | https://ai.google.dev/pricing | Still not listed on official AI Studio page this run either; existing prices retained, not re-verified. |-| Google | gemini-3.6-flash | Input $0.75/MTok, Output $3.75/MTok, Cache read $0.075/MTok (all through Dec 31, 2026; reverts to $1.50/$7.50/$0.15 from Jan 1, 2027) | Yes | No large-context tier | Yes | Updated | https://ai.google.dev/pricing https://ai.google.dev/gemini-api/docs/pricing | **Price cut found 2026-08-14.** Two independent targeted verbatim fetches (both `ai.google.dev/pricing` and `ai.google.dev/gemini-api/docs/pricing`, each explicitly separating Free/Paid columns) confirm Google introduced introductory-style pricing for this model: Paid tier is now $0.75/MTok input, $3.75/MTok output, $0.075/MTok cache read (10% ratio preserved) "through December 31, 2026", stepping up to $1.50/$7.50/$0.15 (the previously-confirmed and still-current file price before this run) "starting January 1, 2027". Since Langfuse's schema has no time-based tiering, the file now holds the current discounted price — **revert to $1.50/$7.50/$0.15 on or after 2027-01-01.** Grounding/web-search-queries price ($14/1,000 requests = 0.014/query) unchanged and confirmed via a dedicated grounding-pricing fetch to apply uniformly across Gemini 3.x models including this one. |-| Google | gemini-3.7-flash | Input $0.75/MTok, Output $3.75/MTok, Cache read $0.075/MTok (all through Dec 31, 2026; reverts to $1.50/$7.50/$0.15 from Jan 1, 2027) | Yes | No large-context tier | Yes | Added | https://ai.google.dev/pricing https://ai.google.dev/gemini-api/docs/pricing https://ai.google.dev/gemini-api/docs/models | **New model found 2026-08-14.** `gemini-3.7-flash` is confirmed via the official Gemini models page as a "New Stable" GA release, described as "Our latest and most capable Flash model, built for complex coding, agentic workflows, and reliable multi-step execution" — the successor to `gemini-3.6-flash` ("previous-generation Flash model"). It launched at the exact same current promotional price as `gemini-3.6-flash` (see that row) and shares the same Jan 1, 2027 step-up. Added to the pricing file mirroring the `gemini-3.6-flash` key set (input/output/cache-read aliases plus `grounding_queries`/`web_search_queries` at 0.014/query) and to `vertexAIModels`/`googleAIStudioModels` in `types.ts` (not as the first entry). matchPattern: `(?i)^(google(ai)?\/)?(gemini-3.7-flash)$`. |+| Google | gemini-3.6-flash | Input $0.75/MTok, Output $3.75/MTok, Cache read $0.075/MTok (all through Dec 31, 2026; reverts to $1.50/$7.50/$0.15 from Jan 1, 2027) | Yes | No large-context tier | Yes | None | https://ai.google.dev/pricing https://ai.google.dev/gemini-api/docs/pricing | Re-confirmed unchanged 2026-08-21 (still on introductory pricing). **Price cut found 2026-08-14.** Two independent targeted verbatim fetches (both `ai.google.dev/pricing` and `ai.google.dev/gemini-api/docs/pricing`, each explicitly separating Free/Paid columns) confirm Google introduced introductory-style pricing for this model: Paid tier is now $0.75/MTok input, $3.75/MTok output, $0.075/MTok cache read (10% ratio preserved) "through December 31, 2026", stepping up to $1.50/$7.50/$0.15 (the previously-confirmed and still-current file price before this run) "starting January 1, 2027". Since Langfuse's schema has no time-based tiering, the file now holds the current discounted price — **revert to $1.50/$7.50/$0.15 on or after 2027-01-01.** Grounding/web-search-queries price ($14/1,000 requests = 0.014/query) unchanged and confirmed via a dedicated grounding-pricing fetch to apply uniformly across Gemini 3.x models including this one. |+| Google | gemini-3.7-flash | Input $0.75/MTok, Output $3.75/MTok, Cache read $0.075/MTok (all through Dec 31, 2026; reverts to $1.50/$7.50/$0.15 from Jan 1, 2027) | Yes | No large-context tier | Yes | None | https://ai.google.dev/pricing https://ai.google.dev/gemini-api/docs/pricing https://ai.google.dev/gemini-api/docs/models | Re-confirmed unchanged 2026-08-21 (still on introductory pricing). **New model found 2026-08-14.** `gemini-3.7-flash` is confirmed via the official Gemini models page as a "New Stable" GA release, described as "Our latest and most capable Flash model, built for complex coding, agentic workflows, and reliable multi-step execution" — the successor to `gemini-3.6-flash` ("previous-generation Flash model"). It launched at the exact same current promotional price as `gemini-3.6-flash` (see that row) and shares the same Jan 1, 2027 step-up. Added to the pricing file mirroring the `gemini-3.6-flash` key set (input/output/cache-read aliases plus `grounding_queries`/`web_search_queries` at 0.014/query) and to `vertexAIModels`/`googleAIStudioModels` in `types.ts` (not as the first entry). matchPattern: `(?i)^(google(ai)?\/)?(gemini-3.7-flash)$`. || Google | gemini-2.0-flash | Input $0.10/MTok, Output $0.40/MTok | No | Deprecated (shut down June 1, 2026) | Not applicable | None | https://ai.google.dev/pricing | Not re-verified this run; retained for backward compatibility. || Google | gemini-2.0-flash-001 | Same as gemini-2.0-flash | No | Deprecated (shut down June 1, 2026) | Not applicable | None | https://ai.google.dev/pricing | Not re-verified this run; retained for backward compatibility. |-## Unresolved findings (updated 2026-08-14)+## Unresolved findings (updated 2026-08-21)1. **claude-opus-4-1-20250805 retirement** — Deprecated, past its originally stated Aug 52026 retirement date but still listed on the official pricing page as "retired, except@@ -157,3 +165,23 @@ All prices listed as `$X / MTok` (per million tokens). Per-token JSON values: diaudits) since they are long-retired and out of the "flagship text/chat/reasoning model"scope. If a future task explicitly asks to audit embeddings or PaLM-era models, treatthis as unverified starting ground, not confirmed.++9. **New Gemini specialized-modality models confirmed out of scope (found 2026-08-21)** —+ `gemini-omni-flash`, `gemini-3.1-flash-live-preview`, `gemini-3.1-flash-tts-preview`,+ `gemini-3.5-live-translate-preview`, the `veo-3.1-*-generate-preview` pair, the+ `lyria-3-*`/`lyria-realtime-exp` family, and `gemini-robotics-er-2-preview` all appeared+ on the official Gemini models page. Each was individually confirmed via its own+ description to be a video-generation, live/voice-only, text-to-speech,+ speech-to-speech-translation, music-generation, or robotics endpoint — none is a+ general-purpose text/chat model with standard per-token text pricing. No pricing or+ `types.ts` entries were added, per the automated-audit rule to skip modality-specific+ endpoints. See `provider-sources-and-price-keys.md` for the per-model detail. Not truly+ "unresolved" — resolved as out of scope — but listed here so future audits don't+ re-spend a fetch re-investigating the same family unless one of them gains a standard+ text-generation mode.++10. **gemini-3.1-flash-lite-preview and gemini-3-pro-preview still absent (re-checked+ 2026-08-21)** — Both remain missing from `ai.google.dev/pricing` and+ `ai.google.dev/gemini-api/docs/pricing`, and `gemini-3-pro-preview` also does not appear+ on `ai.google.dev/gemini-api/docs/models`. Same status as finding #4; existing file+ prices retained without fresh confirmation for a second consecutive run.diff --git a/.agents/skills/add-model-price/references/provider-sources-and-price-keys.md b/.agents/skills/add-model-price/references/provider-sources-and-price-keys.mdindex d03476d656..3fddfefbdc 100644--- a/.agents/skills/add-model-price/references/provider-sources-and-price-keys.md+++ b/.agents/skills/add-model-price/references/provider-sources-and-price-keys.md@@ -173,8 +173,7 @@ Always fetch pricing from the provider's official docs before editing.Pricing: $1.75/MTok input, $0.175/MTok cached input, $14.00/MTok output. Context window:400k tokens; max output 128k tokens. No large-context tier. No date-stamped snapshot atlaunch. Standard OpenAI matchPattern: `(?i)^(openai\/)?(gpt-5.3-codex)$`. Added to pricing- file and `openAIModels` in July 27 2026 audit. Official sources:- `https://developers.openai.com/api/docs/pricing` and+file and `openAIModels`in July 27 2026 audit. Official sources:`https://developers.openai.com/api/docs/pricing` and`https://developers.openai.com/api/docs/models/gpt-5.3-codex`.- **GPT-5.6 Terra / Luna price cut (found July 31 2026)** — OpenAI lowered pricing for`gpt-5.6-terra` and `gpt-5.6-luna` sometime between the July 27 and July 31 2026 audits;@@ -237,7 +236,7 @@ Always fetch pricing from the provider's official docs before editing.condition can never legitimately fire for this model. The tier was removed; the model nowhas only the Standard tier, matching the precedent set by `claude-haiku-4-5-20251001`(also a 200k-context model with no Large Context tier). If a future model is documented- with a *soft* extended-context cap that bills at a premium rate past a threshold below its+ with a _soft_ extended-context cap that bills at a premium rate past a threshold below itshard context-window limit, that would justify a real tier — verify the hard context-windowsize first before trusting an existing Large Context tier on a non-1M-context Claude model.- **AWS Bedrock "Claude 3.5 Sonnet (Public Extended Access)" pricing confirmed real but not@@ -258,6 +257,28 @@ Always fetch pricing from the provider's official docs before editing.alone. Do not add a second pricing entry for this SKU; it would create an unresolvablematchPattern collision with the existing entry. Leave as a documented, confirmedlimitation rather than an open question in future audits.+- **Gemini specialized-modality model wave (found August 21 2026, out of scope)** — The+ official Gemini models page (`ai.google.dev/gemini-api/docs/models`) now lists several+ new model IDs beyond `gemini-3.7-flash`: `gemini-omni-flash` ("Fast, conversational video+ generation and editing... turn text and images into video"), `gemini-3.1-flash-live-preview`+ ("Live API model for real-time dialogue and voice-first AI applications"),+ `gemini-3.1-flash-tts-preview` ("Powerful, low-latency speech generation"),+ `gemini-3.5-live-translate-preview` ("real-time speech to speech translation"), plus+ `veo-3.1-generate-preview`/`veo-3.1-lite-generate-preview` (video), `lyria-3-pro-preview`/+ `lyria-3-clip-preview`/`lyria-realtime-exp` (music), and `gemini-robotics-er-2-preview`+ (robotics). A targeted fetch of each model's description confirmed none is a+ general-purpose text/chat model with standard per-token text pricing — they are video+ generation, live/voice-only, text-to-speech, speech-to-speech translation, music+ generation, and robotics endpoints. Per the automated-audit skip rule for+ modality-specific endpoints, none were added to the pricing file or `types.ts`. Future+ audits do not need to re-investigate this family unless one of them gains a standard+ text-generation mode with its own per-token text pricing.+- **gpt-5-chat-latest confirmed again (August 21 2026)** — Re-fetched+ `https://developers.openai.com/api/docs/models/gpt-5-chat-latest` directly (it is absent+ from the aggregate standard-pricing-table dump, consistent with every prior audit).+ Confirmed unchanged: $1.25/MTok input, $0.125/MTok cached input, $10/MTok output, 128,000+ token context window, no large-context tier. Matches the file exactly+ (id `8ba72ee3-ebe8-4110-a614-bf81094447e5`).- **OpenAI base-model vs. fine-tuning-legacy price mixups (fixed August 7 2026)** — TheOpenAI pricing page lists some base model names in two different tables: the "Standard"table (bare inference pricing, what a `matchPattern` with no `ft:` prefix should use) and@@ -277,6 +298,47 @@ Always fetch pricing from the provider's official docs before editing.applying it to the bare (non-`ft:`) entry — a summarizer can silently pick either tablewhen both rows share the same model name.+- **Premium speed-tier pricing (`service_tier`/`speed`) is documented for far more+ models than the initial rollout covered (implemented 2026-08-20)** — The+ `model_parameters` tier-condition mechanism landed in PR #16204 (2026-08-18) and was+ used to add a "Fast mode" tier (`service_tier` in `["fast","priority"]`) to exactly+ four OpenAI entries: `gpt-5.5-2026-04-23`, `gpt-5.6-sol`, `gpt-5.6-terra`, and+ `gpt-5.6-luna`. Two independent `developers.openai.com/api/docs/pricing` fetches this+ run (one broad, one asking to quote the "Fast mode" table verbatim, plus a request to+ quote the separate "Flex" table verbatim) confirm OpenAI documents official Fast mode+ and Flex processing prices for many more models:+ - **Fast mode** (`service_tier: "fast"` or `"priority"`; "Priority processing" was+ renamed "Fast mode" on 2026-07-30, both values still accepted) — confirmed+ per-MTok short-context prices (input / cached input / output; cache writes only+ where shown): `gpt-5.4` $5.00/$0.50/$30.00, `gpt-5.4-mini` $1.50/$0.15/$9.00,+ `gpt-5.2` $3.50/$0.35/$28.00, `gpt-5.1` $2.50/$0.25/$20.00, `gpt-5` $2.50/$0.25/$20.00,+ `gpt-5-mini` $0.45/$0.045/$3.60, `gpt-4.1` $3.50/$0.875/$14.00, `gpt-4.1-mini`+ $0.70/$0.175/$2.80, `gpt-4.1-nano` $0.20/$0.05/$0.80, `gpt-4o` $4.25/$2.125/$17.00,+ `gpt-4o-2024-05-13` $8.75/—/$26.25, `gpt-4o-mini` $0.25/$0.125/$1.00, `o3`+ $3.50/$0.875/$14.00, `o4-mini` $2.00/$0.50/$8.00. (`gpt-5.5` Fast mode price was+ already re-confirmed as unchanged at $12.50/$1.25/$75.00 in the pricing file.)+ - **Flex** (`service_tier: "flex"`, a discount tier, roughly half of standard) —+ confirmed for `gpt-5.6-sol` $2.50/$0.25/$15.00, `gpt-5.6-terra` $1.00/$0.10/$6.00,+ `gpt-5.6-luna` $0.10/$0.01/$0.60, `gpt-5.5` $2.50/$0.25/$15.00, `gpt-5.4`+ $1.25/$0.13/$7.50, plus `gpt-5.4-mini`, `gpt-5.4-nano`, `gpt-5.4-pro`, `gpt-5.2`,+ `gpt-5.1`, `gpt-5`, `gpt-5-mini`, `gpt-5-nano`, `o3`, and `o4-mini` (prices seen but+ not individually re-quoted during the audit; all were re-read from the live table+ before implementation). The pricing file now represents these with+ `modelParameters.service_tier in ["flex"]`.+ - **Anthropic has the same class of gap**: the pricing page's "Fast mode pricing"+ section documents Claude Opus 5 / Claude Opus 4.8 Fast mode at $10/$50 per MTok+ input/output (`speed: "fast"` request parameter), but neither `claude-opus-5` nor+ `claude-opus-4-8` has a Fast-mode tier in the pricing file (both are single-tier+ `Standard`-only entries before the follow-up). The generated Anthropic Python SDK's+ beta `MessageCreateParamsBase` confirms the request field is `speed`, with values+ `"standard" | "fast"`; Anthropic's separate `service_tier` field controls capacity+ and is not the Fast-mode discriminator. The pricing file therefore matches+ `modelParameters.speed in ["fast"]`.+ - The 2026-08-20 follow-up added every Fast and Flex tier listed above to alias and+ dated-snapshot entries, plus combined Flex/large-context tiers where the documented+ > 272K multiplier applies. It also added the two Anthropic Fast-mode tiers, including+ > the documented prompt-cache multipliers.+Capture:1. Base input token price per million tokensdiff --git a/packages/shared/src/server/repositories/events.ts b/packages/shared/src/server/repositories/events.tsindex 13e665f029..5ad152df4c 100644--- a/packages/shared/src/server/repositories/events.ts+++ b/packages/shared/src/server/repositories/events.ts@@ -269,7 +269,9 @@ async function enrichObservationsWithModelData(OR: [{ projectId: projectId }, { projectId: null }],},include: {- Price: true,+ Price: {+ where: { pricingTier: { isDefault: true } },+ },},}): [];diff --git a/packages/shared/src/server/repositories/observations.ts b/packages/shared/src/server/repositories/observations.tsindex 9d945ec378..986e422697 100644--- a/packages/shared/src/server/repositories/observations.ts+++ b/packages/shared/src/server/repositories/observations.ts@@ -597,7 +597,9 @@ export const getObservationsTableWithModelData = async (OR: [{ projectId: opts.projectId }, { projectId: null }],},include: {- Price: true,+ Price: {+ where: { pricingTier: { isDefault: true } },+ },},}): [],diff --git a/packages/shared/src/server/repositories/observations_converters.ts b/packages/shared/src/server/repositories/observations_converters.tsindex 07bab02736..fd5abc1f9d 100644--- a/packages/shared/src/server/repositories/observations_converters.ts+++ b/packages/shared/src/server/repositories/observations_converters.ts@@ -48,7 +48,9 @@ export const createModelCache = (projectId: string) => {OR: [{ projectId }, { projectId: null }],},include: {- Price: true,+ Price: {+ where: { pricingTier: { isDefault: true } },+ },},});diff --git a/web/src/pages/api/public/observations/[observationId].ts b/web/src/pages/api/public/observations/[observationId].tsindex b191d821f9..84e5b3350e 100644--- a/web/src/pages/api/public/observations/[observationId].ts+++ b/web/src/pages/api/public/observations/[observationId].ts@@ -70,7 +70,9 @@ export default withMiddlewares(],},include: {- Price: true,+ Price: {+ where: { pricingTier: { isDefault: true } },+ },},orderBy: {projectId: {diff --git a/web/src/pages/api/public/observations/index.ts b/web/src/pages/api/public/observations/index.tsindex b5c808c09b..66cda58f3d 100644--- a/web/src/pages/api/public/observations/index.ts+++ b/web/src/pages/api/public/observations/index.ts@@ -95,7 +95,9 @@ export default withMiddlewares(],},include: {- Price: true,+ Price: {+ where: { pricingTier: { isDefault: true } },+ },},}): [];diff --git a/web/src/pages/api/public/traces/[traceId].ts b/web/src/pages/api/public/traces/[traceId].tsindex a604336e1d..b1fddf60f9 100644--- a/web/src/pages/api/public/traces/[traceId].ts+++ b/web/src/pages/api/public/traces/[traceId].ts@@ -117,7 +117,9 @@ export default withMiddlewares(],},include: {- Price: true,+ Price: {+ where: { pricingTier: { isDefault: true } },+ },},}): [];diff --git a/worker/src/__tests__/pricing-tier-matcher.test.ts b/worker/src/__tests__/pricing-tier-matcher.test.tsindex b18ef7f541..0c078515bb 100644--- a/worker/src/__tests__/pricing-tier-matcher.test.ts+++ b/worker/src/__tests__/pricing-tier-matcher.test.ts@@ -322,7 +322,9 @@ describe("default-model-prices.json", () => {expect(model!.pricingTiers.map((tier) => tier.name)).toEqual(["Standard","Fast mode · Large context (>272K)",+ "Flex · Large context (>272K)","Fast mode",+ "Flex","Large Context (>272K)",]);@@ -406,6 +408,22 @@ describe("default-model-prices.json", () => {{ modelParameters: { service_tier: "fast" } },)?.pricingTierName,).toBe("Fast mode");++ expect(+ matchPricingTier(+ tiers,+ { input: 1000 },+ { modelParameters: { service_tier: "flex" } },+ )?.pricingTierName,+ ).toBe("Flex");++ expect(+ matchPricingTier(+ tiers,+ { cache_write_tokens: 272001 },+ { modelParameters: { service_tier: "flex" } },+ )?.pricingTierName,+ ).toBe("Flex · Large context (>272K)");});it.each(["fast", "priority"])(@@ -439,6 +457,159 @@ describe("default-model-prices.json", () => {).toBe("Fast mode");},);++ it.each([+ ["gpt-5.4", 5, 30],+ ["gpt-5.4-2026-03-05", 5, 30],+ ["gpt-5.4-mini", 1.5, 9],+ ["gpt-5.4-mini-2026-03-17", 1.5, 9],+ ["gpt-5.2", 3.5, 28],+ ["gpt-5.2-2025-12-11", 3.5, 28],+ ["gpt-5.1", 2.5, 20],+ ["gpt-5.1-2025-11-13", 2.5, 20],+ ["gpt-5", 2.5, 20],+ ["gpt-5-2025-08-07", 2.5, 20],+ ["gpt-5-mini", 0.45, 3.6],+ ["gpt-5-mini-2025-08-07", 0.45, 3.6],+ ["gpt-4.1", 3.5, 14],+ ["gpt-4.1-2025-04-14", 3.5, 14],+ ["gpt-4.1-mini", 0.7, 2.8],+ ["gpt-4.1-mini-2025-04-14", 0.7, 2.8],+ ["gpt-4.1-nano", 0.2, 0.8],+ ["gpt-4.1-nano-2025-04-14", 0.2, 0.8],+ ["gpt-4o", 4.25, 17],+ ["gpt-4o-2024-05-13", 8.75, 26.25],+ ["gpt-4o-2024-08-06", 4.25, 17],+ ["gpt-4o-2024-11-20", 4.25, 17],+ ["gpt-4o-mini", 0.25, 1],+ ["gpt-4o-mini-2024-07-18", 0.25, 1],+ ["o3", 3.5, 14],+ ["o3-2025-04-16", 3.5, 14],+ ["o4-mini", 2, 8],+ ["o4-mini-2025-04-16", 2, 8],+ ])(+ "should price %s Fast mode",+ (modelName, inputPerMillion, outputPerMillion) => {+ const model = defaultModelPrices.find(+ (candidate) => candidate.modelName === modelName,+ );+ const tier = model?.pricingTiers.find(+ (candidate) => candidate.name === "Fast mode",+ );++ expect(tier, modelName).toBeDefined();+ expect(tier?.conditions).toContainEqual({+ source: "model_parameters",+ key: "service_tier",+ operator: "in",+ values: ["fast", "priority"],+ });+ expect(tier?.prices.input).toBeCloseTo(inputPerMillion * 1e-6, 15);+ expect(tier?.prices.output).toBeCloseTo(outputPerMillion * 1e-6, 15);+ },+ );++ it.each([+ ["gpt-5.6-sol", 2.5, 15],+ ["gpt-5.6-terra", 1, 6],+ ["gpt-5.6-luna", 0.1, 0.6],+ ["gpt-5.5-2026-04-23", 2.5, 15],+ ["gpt-5.5-pro-2026-04-23", 15, 90],+ ["gpt-5.4", 1.25, 7.5],+ ["gpt-5.4-2026-03-05", 1.25, 7.5],+ ["gpt-5.4-pro", 15, 90],+ ["gpt-5.4-pro-2026-03-05", 15, 90],+ ["gpt-5.4-mini", 0.375, 2.25],+ ["gpt-5.4-mini-2026-03-17", 0.375, 2.25],+ ["gpt-5.4-nano", 0.1, 0.625],+ ["gpt-5.4-nano-2026-03-17", 0.1, 0.625],+ ["gpt-5.2", 0.875, 7],+ ["gpt-5.2-2025-12-11", 0.875, 7],+ ["gpt-5.1", 0.625, 5],+ ["gpt-5.1-2025-11-13", 0.625, 5],+ ["gpt-5", 0.625, 5],+ ["gpt-5-2025-08-07", 0.625, 5],+ ["gpt-5-mini", 0.125, 1],+ ["gpt-5-mini-2025-08-07", 0.125, 1],+ ["gpt-5-nano", 0.025, 0.2],+ ["gpt-5-nano-2025-08-07", 0.025, 0.2],+ ["o3", 1, 4],+ ["o3-2025-04-16", 1, 4],+ ["o4-mini", 0.55, 2.2],+ ["o4-mini-2025-04-16", 0.55, 2.2],+ ])(+ "should price %s Flex processing",+ (modelName, inputPerMillion, outputPerMillion) => {+ const model = defaultModelPrices.find(+ (candidate) => candidate.modelName === modelName,+ );+ const tier = model?.pricingTiers.find(+ (candidate) => candidate.name === "Flex",+ );++ expect(tier, modelName).toBeDefined();+ expect(tier?.conditions).toEqual([+ {+ source: "model_parameters",+ key: "service_tier",+ operator: "in",+ values: ["flex"],+ },+ ]);+ expect(tier?.prices.input).toBeCloseTo(inputPerMillion * 1e-6, 15);+ expect(tier?.prices.output).toBeCloseTo(outputPerMillion * 1e-6, 15);+ },+ );++ it.each(["claude-opus-5", "claude-opus-4-8"])(+ "should price %s Fast mode using Anthropic's speed parameter",+ (modelName) => {+ const model = defaultModelPrices.find(+ (candidate) => candidate.modelName === modelName,+ );+ const tier = model?.pricingTiers.find(+ (candidate) => candidate.name === "Fast mode",+ );++ expect(tier, modelName).toBeDefined();+ expect(tier?.conditions).toEqual([+ {+ source: "model_parameters",+ key: "speed",+ operator: "in",+ values: ["fast"],+ },+ ]);+ expect(tier?.prices.input).toBe(10e-6);+ expect(tier?.prices.output).toBe(50e-6);+ expect(tier?.prices.input_cache_creation_5m).toBe(12.5e-6);+ expect(tier?.prices.input_cache_creation_1h).toBe(20e-6);+ expect(tier?.prices.input_cache_read).toBe(1e-6);++ const tiers: PricingTierWithPrices[] = model!.pricingTiers.map(+ (candidate) => ({+ id: candidate.id,+ name: candidate.name,+ isDefault: candidate.isDefault,+ priority: candidate.priority,+ conditions: candidate.conditions,+ prices: Object.entries(candidate.prices).map(+ ([usageType, price]) => ({+ usageType,+ price: new Decimal(price),+ }),+ ),+ }),+ );+ expect(+ matchPricingTier(+ tiers,+ { input: 1000 },+ { modelParameters: { speed: "fast" } },+ )?.pricingTierName,+ ).toBe("Fast mode");+ },+ );});describe("validateRegexPattern", () => {diff --git a/worker/src/constants/default-model-prices.json b/worker/src/constants/default-model-prices.jsonindex eae50924bc..4cbc4aa050 100644--- a/worker/src/constants/default-model-prices.json+++ b/worker/src/constants/default-model-prices.json@@ -4,7 +4,7 @@"modelName": "gpt-4o","matchPattern": "(?i)^(openai\/)?(gpt-4o)$","createdAt": "2024-05-13T23:15:07.670Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4o",@@ -19,10 +19,32 @@"priority": 0,"conditions": [],"prices": {- "input": 0.0000025,- "input_cached_tokens": 0.00000125,- "input_cache_read": 0.00000125,- "output": 0.00001+ "input": 2.5e-6,+ "input_cached_tokens": 1.25e-6,+ "input_cache_read": 1.25e-6,+ "output": 10e-6,+ "cache_read_input_tokens": 1.25e-6+ }+ },+ {+ "id": "ba3b2121-9c68-4333-af3b-26f98fbab7ee",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 4.25e-6,+ "input_cached_tokens": 2.125e-6,+ "input_cache_read": 2.125e-6,+ "output": 17e-6,+ "cache_read_input_tokens": 2.125e-6}}]@@ -32,7 +54,7 @@"modelName": "gpt-4o-2024-05-13","matchPattern": "(?i)^(openai\/)?(gpt-4o-2024-05-13)$","createdAt": "2024-05-13T23:15:07.670Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4o-2024-05-13",@@ -47,8 +69,26 @@"priority": 0,"conditions": [],"prices": {- "input": 0.000005,- "output": 0.000015+ "input": 5e-6,+ "output": 15e-6+ }+ },+ {+ "id": "62d76894-9b2d-444a-985e-6802a1c730aa",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 8.75e-6,+ "output": 26.25e-6}}]@@ -1569,7 +1609,7 @@"modelName": "gpt-4o-mini","matchPattern": "(?i)^(openai\/)?(gpt-4o-mini)$","createdAt": "2024-07-18T17:56:09.591Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4o",@@ -1584,10 +1624,32 @@"priority": 0,"conditions": [],"prices": {- "input": 1.5e-7,- "output": 6e-7,- "input_cached_tokens": 7.5e-8,- "input_cache_read": 7.5e-8+ "input": 0.15e-6,+ "output": 0.6e-6,+ "input_cached_tokens": 0.075e-6,+ "input_cache_read": 0.075e-6,+ "cache_read_input_tokens": 0.075e-6+ }+ },+ {+ "id": "6a318ead-09a7-49cd-9cfd-e3c799b3716a",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 0.25e-6,+ "output": 1e-6,+ "input_cached_tokens": 0.125e-6,+ "input_cache_read": 0.125e-6,+ "cache_read_input_tokens": 0.125e-6}}]@@ -1597,7 +1659,7 @@"modelName": "gpt-4o-mini-2024-07-18","matchPattern": "(?i)^(openai\/)?(gpt-4o-mini-2024-07-18)$","createdAt": "2024-07-18T17:56:09.591Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4o",@@ -1612,10 +1674,32 @@"priority": 0,"conditions": [],"prices": {- "input": 1.5e-7,- "input_cached_tokens": 7.5e-8,- "input_cache_read": 7.5e-8,- "output": 6e-7+ "input": 0.15e-6,+ "input_cached_tokens": 0.075e-6,+ "input_cache_read": 0.075e-6,+ "output": 0.6e-6,+ "cache_read_input_tokens": 0.075e-6+ }+ },+ {+ "id": "204f0b4b-5605-4f63-bec5-3d62c2b810b4",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 0.25e-6,+ "input_cached_tokens": 0.125e-6,+ "input_cache_read": 0.125e-6,+ "output": 1e-6,+ "cache_read_input_tokens": 0.125e-6}}]@@ -1625,7 +1709,7 @@"modelName": "gpt-4o-2024-08-06","matchPattern": "(?i)^(openai\/)?(gpt-4o-2024-08-06)$","createdAt": "2024-08-07T11:54:31.298Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4o",@@ -1640,10 +1724,32 @@"priority": 0,"conditions": [],"prices": {- "input": 0.0000025,- "input_cached_tokens": 0.00000125,- "input_cache_read": 0.00000125,- "output": 0.00001+ "input": 2.5e-6,+ "input_cached_tokens": 1.25e-6,+ "input_cache_read": 1.25e-6,+ "output": 10e-6,+ "cache_read_input_tokens": 1.25e-6+ }+ },+ {+ "id": "0e6f6dac-20cc-4ae2-bfab-90070ac2de4a",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 4.25e-6,+ "input_cached_tokens": 2.125e-6,+ "input_cache_read": 2.125e-6,+ "output": 17e-6,+ "cache_read_input_tokens": 2.125e-6}}]@@ -1895,7 +2001,7 @@"modelName": "gpt-4o-2024-11-20","matchPattern": "(?i)^(openai\/)?(gpt-4o-2024-11-20)$","createdAt": "2024-12-03T10:06:12.000Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4o",@@ -1910,10 +2016,32 @@"priority": 0,"conditions": [],"prices": {- "input": 0.0000025,- "input_cached_tokens": 0.00000125,- "input_cache_read": 0.00000125,- "output": 0.00001+ "input": 2.5e-6,+ "input_cached_tokens": 1.25e-6,+ "input_cache_read": 1.25e-6,+ "output": 10e-6,+ "cache_read_input_tokens": 1.25e-6+ }+ },+ {+ "id": "eb22e14f-7474-422a-aac3-2d0f722ceba0",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 4.25e-6,+ "input_cached_tokens": 2.125e-6,+ "input_cache_read": 2.125e-6,+ "output": 17e-6,+ "cache_read_input_tokens": 2.125e-6}}]@@ -2265,7 +2393,7 @@"modelName": "gpt-4.1","matchPattern": "(?i)^(openai\/)?(gpt-4.1)$","createdAt": "2025-04-15T10:26:54.132Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -2280,11 +2408,34 @@"priority": 0,"conditions": [],"prices": {- "input": 0.000002,- "input_cached_tokens": 5e-7,- "input_cached_text_tokens": 5e-7,- "input_cache_read": 5e-7,- "output": 0.000008+ "input": 2e-6,+ "input_cached_tokens": 0.5e-6,+ "input_cached_text_tokens": 0.5e-6,+ "input_cache_read": 0.5e-6,+ "output": 8e-6,+ "cache_read_input_tokens": 0.5e-6+ }+ },+ {+ "id": "757639cc-73fa-49c5-adfd-686bbd0e9dbe",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 3.5e-6,+ "input_cached_tokens": 0.875e-6,+ "input_cached_text_tokens": 0.875e-6,+ "input_cache_read": 0.875e-6,+ "output": 14e-6,+ "cache_read_input_tokens": 0.875e-6}}]@@ -2294,7 +2445,7 @@"modelName": "gpt-4.1-2025-04-14","matchPattern": "(?i)^(openai\/)?(gpt-4.1-2025-04-14)$","createdAt": "2025-04-15T10:26:54.132Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -2309,11 +2460,34 @@"priority": 0,"conditions": [],"prices": {- "input": 0.000002,- "input_cached_tokens": 5e-7,- "input_cached_text_tokens": 5e-7,- "input_cache_read": 5e-7,- "output": 0.000008+ "input": 2e-6,+ "input_cached_tokens": 0.5e-6,+ "input_cached_text_tokens": 0.5e-6,+ "input_cache_read": 0.5e-6,+ "output": 8e-6,+ "cache_read_input_tokens": 0.5e-6+ }+ },+ {+ "id": "ef4f4063-bd8f-441c-b2a2-6771f404212e",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 3.5e-6,+ "input_cached_tokens": 0.875e-6,+ "input_cached_text_tokens": 0.875e-6,+ "input_cache_read": 0.875e-6,+ "output": 14e-6,+ "cache_read_input_tokens": 0.875e-6}}]@@ -2323,7 +2497,7 @@"modelName": "gpt-4.1-mini-2025-04-14","matchPattern": "(?i)^(openai\/)?(gpt-4.1-mini-2025-04-14)$","createdAt": "2025-04-15T10:26:54.132Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -2338,11 +2512,34 @@"priority": 0,"conditions": [],"prices": {- "input": 4e-7,- "input_cached_tokens": 1e-7,- "input_cached_text_tokens": 1e-7,- "input_cache_read": 1e-7,- "output": 0.0000016+ "input": 0.4e-6,+ "input_cached_tokens": 0.1e-6,+ "input_cached_text_tokens": 0.1e-6,+ "input_cache_read": 0.1e-6,+ "output": 1.6e-6,+ "cache_read_input_tokens": 0.1e-6+ }+ },+ {+ "id": "94d78fcb-36e0-4dd7-b7ae-7d43e8dd278b",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 0.7e-6,+ "input_cached_tokens": 0.175e-6,+ "input_cached_text_tokens": 0.175e-6,+ "input_cache_read": 0.175e-6,+ "output": 2.8e-6,+ "cache_read_input_tokens": 0.175e-6}}]@@ -2352,7 +2549,7 @@"modelName": "gpt-4.1-nano-2025-04-14","matchPattern": "(?i)^(openai\/)?(gpt-4.1-nano-2025-04-14)$","createdAt": "2025-04-15T10:26:54.132Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -2367,11 +2564,34 @@"priority": 0,"conditions": [],"prices": {- "input": 1e-7,- "input_cached_tokens": 2.5e-8,- "input_cached_text_tokens": 2.5e-8,- "input_cache_read": 2.5e-8,- "output": 4e-7+ "input": 0.1e-6,+ "input_cached_tokens": 0.025e-6,+ "input_cached_text_tokens": 0.025e-6,+ "input_cache_read": 0.025e-6,+ "output": 0.4e-6,+ "cache_read_input_tokens": 0.025e-6+ }+ },+ {+ "id": "38ecc517-b487-45d0-abea-7dbbf1a706a6",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 0.2e-6,+ "input_cached_tokens": 0.05e-6,+ "input_cached_text_tokens": 0.05e-6,+ "input_cache_read": 0.05e-6,+ "output": 0.8e-6,+ "cache_read_input_tokens": 0.05e-6}}]@@ -2381,7 +2601,7 @@"modelName": "o3","matchPattern": "(?i)^(openai\/)?(o3)$","createdAt": "2025-04-16T23:26:54.132Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","pricingTiers": [{"id": "cm7wmny967124dhjtb95w8f81_tier_default",@@ -2390,12 +2610,62 @@"priority": 0,"conditions": [],"prices": {- "input": 0.000002,- "input_cached_tokens": 5e-7,- "input_cache_read": 5e-7,- "output": 0.000008,- "output_reasoning_tokens": 0.000008,- "output_reasoning": 0.000008+ "input": 2e-6,+ "input_cached_tokens": 0.5e-6,+ "input_cache_read": 0.5e-6,+ "output": 8e-6,+ "output_reasoning_tokens": 8e-6,+ "output_reasoning": 8e-6,+ "cache_read_input_tokens": 0.5e-6,+ "reasoning_tokens": 8e-6+ }+ },+ {+ "id": "ad862208-0cc4-4ad6-868d-363fbaace2b2",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 3.5e-6,+ "input_cached_tokens": 0.875e-6,+ "input_cache_read": 0.875e-6,+ "output": 14e-6,+ "output_reasoning_tokens": 14e-6,+ "output_reasoning": 14e-6,+ "cache_read_input_tokens": 0.875e-6,+ "reasoning_tokens": 14e-6+ }+ },+ {+ "id": "0c5034e2-181e-4942-bd65-4a210bfd48dd",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 1e-6,+ "input_cached_tokens": 0.25e-6,+ "input_cache_read": 0.25e-6,+ "output": 4e-6,+ "output_reasoning_tokens": 4e-6,+ "output_reasoning": 4e-6,+ "cache_read_input_tokens": 0.25e-6,+ "reasoning_tokens": 4e-6}}]@@ -2405,7 +2675,7 @@"modelName": "o3-2025-04-16","matchPattern": "(?i)^(openai\/)?(o3-2025-04-16)$","createdAt": "2025-04-16T23:26:54.132Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","pricingTiers": [{"id": "cm7wopq3327124dhjtb95w8f81_tier_default",@@ -2414,22 +2684,72 @@"priority": 0,"conditions": [],"prices": {- "input": 0.000002,- "input_cached_tokens": 5e-7,- "input_cache_read": 5e-7,- "output": 0.000008,- "output_reasoning_tokens": 0.000008,- "output_reasoning": 0.000008+ "input": 2e-6,+ "input_cached_tokens": 0.5e-6,+ "input_cache_read": 0.5e-6,+ "output": 8e-6,+ "output_reasoning_tokens": 8e-6,+ "output_reasoning": 8e-6,+ "cache_read_input_tokens": 0.5e-6,+ "reasoning_tokens": 8e-6}- }- ]- },- {+ },+ {+ "id": "0aa1a211-4af7-43a8-86e2-d193beda27a1",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 3.5e-6,+ "input_cached_tokens": 0.875e-6,+ "input_cache_read": 0.875e-6,+ "output": 14e-6,+ "output_reasoning_tokens": 14e-6,+ "output_reasoning": 14e-6,+ "cache_read_input_tokens": 0.875e-6,+ "reasoning_tokens": 14e-6+ }+ },+ {+ "id": "54114e6b-05a0-4e6c-92e6-b5683245c1dc",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 1e-6,+ "input_cached_tokens": 0.25e-6,+ "input_cache_read": 0.25e-6,+ "output": 4e-6,+ "output_reasoning_tokens": 4e-6,+ "output_reasoning": 4e-6,+ "cache_read_input_tokens": 0.25e-6,+ "reasoning_tokens": 4e-6+ }+ }+ ]+ },+ {"id": "cm7wqrs1327124dhjtb95w8f81","modelName": "o4-mini","matchPattern": "(?i)^(openai\/)?(o4-mini)$","createdAt": "2025-04-16T23:26:54.132Z",- "updatedAt": "2026-06-27T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","pricingTiers": [{"id": "cm7wqrs1327124dhjtb95w8f81_tier_default",@@ -2438,12 +2758,62 @@"priority": 0,"conditions": [],"prices": {- "input": 0.0000011,- "input_cached_tokens": 2.75e-7,- "input_cache_read": 2.75e-7,- "output": 0.0000044,- "output_reasoning_tokens": 0.0000044,- "output_reasoning": 0.0000044+ "input": 1.1e-6,+ "input_cached_tokens": 0.275e-6,+ "input_cache_read": 0.275e-6,+ "output": 4.4e-6,+ "output_reasoning_tokens": 4.4e-6,+ "output_reasoning": 4.4e-6,+ "cache_read_input_tokens": 0.275e-6,+ "reasoning_tokens": 4.4e-6+ }+ },+ {+ "id": "1ddaa9b4-df3b-4ab2-9969-62b263974086",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 2e-6,+ "input_cached_tokens": 0.5e-6,+ "input_cache_read": 0.5e-6,+ "output": 8e-6,+ "output_reasoning_tokens": 8e-6,+ "output_reasoning": 8e-6,+ "cache_read_input_tokens": 0.5e-6,+ "reasoning_tokens": 8e-6+ }+ },+ {+ "id": "02bf5921-4756-4997-9eee-232a7d6e0567",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.55e-6,+ "input_cached_tokens": 0.138e-6,+ "input_cache_read": 0.138e-6,+ "output": 2.2e-6,+ "output_reasoning_tokens": 2.2e-6,+ "output_reasoning": 2.2e-6,+ "cache_read_input_tokens": 0.138e-6,+ "reasoning_tokens": 2.2e-6}}]@@ -2453,7 +2823,7 @@"modelName": "o4-mini-2025-04-16","matchPattern": "(?i)^(openai\/)?(o4-mini-2025-04-16)$","createdAt": "2025-04-16T23:26:54.132Z",- "updatedAt": "2026-06-27T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","pricingTiers": [{"id": "cm7zqrs1327124dhjtb95w8f82_tier_default",@@ -2462,12 +2832,62 @@"priority": 0,"conditions": [],"prices": {- "input": 0.0000011,- "input_cached_tokens": 2.75e-7,- "input_cache_read": 2.75e-7,- "output": 0.0000044,- "output_reasoning_tokens": 0.0000044,- "output_reasoning": 0.0000044+ "input": 1.1e-6,+ "input_cached_tokens": 0.275e-6,+ "input_cache_read": 0.275e-6,+ "output": 4.4e-6,+ "output_reasoning_tokens": 4.4e-6,+ "output_reasoning": 4.4e-6,+ "cache_read_input_tokens": 0.275e-6,+ "reasoning_tokens": 4.4e-6+ }+ },+ {+ "id": "ccb75c3f-8fd7-487c-a5f0-1fefae485dd3",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 2e-6,+ "input_cached_tokens": 0.5e-6,+ "input_cache_read": 0.5e-6,+ "output": 8e-6,+ "output_reasoning_tokens": 8e-6,+ "output_reasoning": 8e-6,+ "cache_read_input_tokens": 0.5e-6,+ "reasoning_tokens": 8e-6+ }+ },+ {+ "id": "5be139db-2dfd-49d9-ae69-40a9a0018be1",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.55e-6,+ "input_cached_tokens": 0.138e-6,+ "input_cache_read": 0.138e-6,+ "output": 2.2e-6,+ "output_reasoning_tokens": 2.2e-6,+ "output_reasoning": 2.2e-6,+ "cache_read_input_tokens": 0.138e-6,+ "reasoning_tokens": 2.2e-6}}]@@ -2517,7 +2937,7 @@"modelName": "gpt-4.1-nano","matchPattern": "(?i)^(openai\/)?(gpt-4.1-nano)$","createdAt": "2025-04-22T10:11:35.241Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -2532,11 +2952,34 @@"priority": 0,"conditions": [],"prices": {- "input": 1e-7,- "input_cached_tokens": 2.5e-8,- "input_cached_text_tokens": 2.5e-8,- "input_cache_read": 2.5e-8,- "output": 4e-7+ "input": 0.1e-6,+ "input_cached_tokens": 0.025e-6,+ "input_cached_text_tokens": 0.025e-6,+ "input_cache_read": 0.025e-6,+ "output": 0.4e-6,+ "cache_read_input_tokens": 0.025e-6+ }+ },+ {+ "id": "27947175-4b36-475d-ac36-44f69fbf1eac",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 0.2e-6,+ "input_cached_tokens": 0.05e-6,+ "input_cached_text_tokens": 0.05e-6,+ "input_cache_read": 0.05e-6,+ "output": 0.8e-6,+ "cache_read_input_tokens": 0.05e-6}}]@@ -2546,7 +2989,7 @@"modelName": "gpt-4.1-mini","matchPattern": "(?i)^(openai\/)?(gpt-4.1-mini)$","createdAt": "2025-04-22T10:11:35.241Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -2561,11 +3004,34 @@"priority": 0,"conditions": [],"prices": {- "input": 4e-7,- "input_cached_tokens": 1e-7,- "input_cached_text_tokens": 1e-7,- "input_cache_read": 1e-7,- "output": 0.0000016+ "input": 0.4e-6,+ "input_cached_tokens": 0.1e-6,+ "input_cached_text_tokens": 0.1e-6,+ "input_cache_read": 0.1e-6,+ "output": 1.6e-6,+ "cache_read_input_tokens": 0.1e-6+ }+ },+ {+ "id": "e747c083-dae8-4085-a1e2-e46c16e39498",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 0.7e-6,+ "input_cached_tokens": 0.175e-6,+ "input_cached_text_tokens": 0.175e-6,+ "input_cache_read": 0.175e-6,+ "output": 2.8e-6,+ "cache_read_input_tokens": 0.175e-6}}]@@ -2883,7 +3349,7 @@"modelName": "gpt-5","matchPattern": "(?i)^(openai\/)?(gpt-5)$","createdAt": "2025-08-07T16:00:00.000Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -2898,12 +3364,62 @@"priority": 0,"conditions": [],"prices": {- "input": 0.00000125,- "input_cached_tokens": 1.25e-7,- "output": 0.00001,- "input_cache_read": 1.25e-7,- "output_reasoning_tokens": 0.00001,- "output_reasoning": 0.00001+ "input": 1.25e-6,+ "input_cached_tokens": 0.125e-6,+ "output": 10e-6,+ "input_cache_read": 0.125e-6,+ "output_reasoning_tokens": 10e-6,+ "output_reasoning": 10e-6,+ "cache_read_input_tokens": 0.125e-6,+ "reasoning_tokens": 10e-6+ }+ },+ {+ "id": "a67e2dee-d22c-4717-8f90-38a38b8a9e0c",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 2.5e-6,+ "input_cached_tokens": 0.25e-6,+ "output": 20e-6,+ "input_cache_read": 0.25e-6,+ "output_reasoning_tokens": 20e-6,+ "output_reasoning": 20e-6,+ "cache_read_input_tokens": 0.25e-6,+ "reasoning_tokens": 20e-6+ }+ },+ {+ "id": "be4042e8-253f-4f20-94a8-2508b0dff73b",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.625e-6,+ "input_cached_tokens": 0.0625e-6,+ "output": 5e-6,+ "input_cache_read": 0.0625e-6,+ "output_reasoning_tokens": 5e-6,+ "output_reasoning": 5e-6,+ "cache_read_input_tokens": 0.0625e-6,+ "reasoning_tokens": 5e-6}}]@@ -2913,7 +3429,7 @@"modelName": "gpt-5-2025-08-07","matchPattern": "(?i)^(openai\/)?(gpt-5-2025-08-07)$","createdAt": "2025-08-11T08:00:00.000Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -2928,12 +3444,62 @@"priority": 0,"conditions": [],"prices": {- "input": 0.00000125,- "input_cached_tokens": 1.25e-7,- "output": 0.00001,- "input_cache_read": 1.25e-7,- "output_reasoning_tokens": 0.00001,- "output_reasoning": 0.00001+ "input": 1.25e-6,+ "input_cached_tokens": 0.125e-6,+ "output": 10e-6,+ "input_cache_read": 0.125e-6,+ "output_reasoning_tokens": 10e-6,+ "output_reasoning": 10e-6,+ "cache_read_input_tokens": 0.125e-6,+ "reasoning_tokens": 10e-6+ }+ },+ {+ "id": "76480c39-b037-42eb-b553-640c7df3a3f1",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 2.5e-6,+ "input_cached_tokens": 0.25e-6,+ "output": 20e-6,+ "input_cache_read": 0.25e-6,+ "output_reasoning_tokens": 20e-6,+ "output_reasoning": 20e-6,+ "cache_read_input_tokens": 0.25e-6,+ "reasoning_tokens": 20e-6+ }+ },+ {+ "id": "5988c80e-a2c3-42b9-8167-ba6c903a4a2b",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.625e-6,+ "input_cached_tokens": 0.0625e-6,+ "output": 5e-6,+ "input_cache_read": 0.0625e-6,+ "output_reasoning_tokens": 5e-6,+ "output_reasoning": 5e-6,+ "cache_read_input_tokens": 0.0625e-6,+ "reasoning_tokens": 5e-6}}]@@ -2943,7 +3509,7 @@"modelName": "gpt-5-mini","matchPattern": "(?i)^(openai\/)?(gpt-5-mini)$","createdAt": "2025-08-07T16:00:00.000Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -2958,27 +3524,77 @@"priority": 0,"conditions": [],"prices": {- "input": 2.5e-7,- "input_cached_tokens": 2.5e-8,- "output": 0.000002,- "input_cache_read": 2.5e-8,- "output_reasoning_tokens": 0.000002,- "output_reasoning": 0.000002+ "input": 0.25e-6,+ "input_cached_tokens": 0.025e-6,+ "output": 2e-6,+ "input_cache_read": 0.025e-6,+ "output_reasoning_tokens": 2e-6,+ "output_reasoning": 2e-6,+ "cache_read_input_tokens": 0.025e-6,+ "reasoning_tokens": 2e-6}- }- ]- },- {- "id": "03b83894-7172-4e1e-8e8b-37d792484efd",- "modelName": "gpt-5-mini-2025-08-07",- "matchPattern": "(?i)^(openai\/)?(gpt-5-mini-2025-08-07)$",- "createdAt": "2025-08-11T08:00:00.000Z",- "updatedAt": "2025-12-12T15:00:06.513Z",- "tokenizerConfig": {- "tokensPerName": 1,- "tokenizerModel": "gpt-4",- "tokensPerMessage": 3- },+ },+ {+ "id": "fa4c5173-323e-41e4-8066-071ac910aa80",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 0.45e-6,+ "input_cached_tokens": 0.045e-6,+ "output": 3.6e-6,+ "input_cache_read": 0.045e-6,+ "output_reasoning_tokens": 3.6e-6,+ "output_reasoning": 3.6e-6,+ "cache_read_input_tokens": 0.045e-6,+ "reasoning_tokens": 3.6e-6+ }+ },+ {+ "id": "5fe8248d-f3f4-49fd-ad76-d3654eac8593",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.125e-6,+ "input_cached_tokens": 0.0125e-6,+ "output": 1e-6,+ "input_cache_read": 0.0125e-6,+ "output_reasoning_tokens": 1e-6,+ "output_reasoning": 1e-6,+ "cache_read_input_tokens": 0.0125e-6,+ "reasoning_tokens": 1e-6+ }+ }+ ]+ },+ {+ "id": "03b83894-7172-4e1e-8e8b-37d792484efd",+ "modelName": "gpt-5-mini-2025-08-07",+ "matchPattern": "(?i)^(openai\/)?(gpt-5-mini-2025-08-07)$",+ "createdAt": "2025-08-11T08:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z",+ "tokenizerConfig": {+ "tokensPerName": 1,+ "tokenizerModel": "gpt-4",+ "tokensPerMessage": 3+ },"tokenizerId": "openai","pricingTiers": [{@@ -2988,12 +3604,62 @@"priority": 0,"conditions": [],"prices": {- "input": 2.5e-7,- "input_cached_tokens": 2.5e-8,- "output": 0.000002,- "input_cache_read": 2.5e-8,- "output_reasoning_tokens": 0.000002,- "output_reasoning": 0.000002+ "input": 0.25e-6,+ "input_cached_tokens": 0.025e-6,+ "output": 2e-6,+ "input_cache_read": 0.025e-6,+ "output_reasoning_tokens": 2e-6,+ "output_reasoning": 2e-6,+ "cache_read_input_tokens": 0.025e-6,+ "reasoning_tokens": 2e-6+ }+ },+ {+ "id": "4c70b81b-e828-459f-80da-8218f7d816d8",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 0.45e-6,+ "input_cached_tokens": 0.045e-6,+ "output": 3.6e-6,+ "input_cache_read": 0.045e-6,+ "output_reasoning_tokens": 3.6e-6,+ "output_reasoning": 3.6e-6,+ "cache_read_input_tokens": 0.045e-6,+ "reasoning_tokens": 3.6e-6+ }+ },+ {+ "id": "810a215a-d0ea-456d-bc19-64f2e28f2d2f",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.125e-6,+ "input_cached_tokens": 0.0125e-6,+ "output": 1e-6,+ "input_cache_read": 0.0125e-6,+ "output_reasoning_tokens": 1e-6,+ "output_reasoning": 1e-6,+ "cache_read_input_tokens": 0.0125e-6,+ "reasoning_tokens": 1e-6}}]@@ -3003,7 +3669,7 @@"modelName": "gpt-5-nano","matchPattern": "(?i)^(openai\/)?(gpt-5-nano)$","createdAt": "2025-08-07T16:00:00.000Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -3018,12 +3684,38 @@"priority": 0,"conditions": [],"prices": {- "input": 5e-8,- "input_cached_tokens": 5e-9,- "output": 4e-7,- "input_cache_read": 5e-9,- "output_reasoning_tokens": 4e-7,- "output_reasoning": 4e-7+ "input": 0.05e-6,+ "input_cached_tokens": 0.005e-6,+ "output": 0.4e-6,+ "input_cache_read": 0.005e-6,+ "output_reasoning_tokens": 0.4e-6,+ "output_reasoning": 0.4e-6,+ "cache_read_input_tokens": 0.005e-6,+ "reasoning_tokens": 0.4e-6+ }+ },+ {+ "id": "74d737d8-ac19-4992-896a-e389bf46842a",+ "name": "Flex",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.025e-6,+ "input_cached_tokens": 0.0025e-6,+ "output": 0.2e-6,+ "input_cache_read": 0.0025e-6,+ "output_reasoning_tokens": 0.2e-6,+ "output_reasoning": 0.2e-6,+ "cache_read_input_tokens": 0.0025e-6,+ "reasoning_tokens": 0.2e-6}}]@@ -3033,7 +3725,7 @@"modelName": "gpt-5-nano-2025-08-07","matchPattern": "(?i)^(openai\/)?(gpt-5-nano-2025-08-07)$","createdAt": "2025-08-11T08:00:00.000Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -3048,12 +3740,38 @@"priority": 0,"conditions": [],"prices": {- "input": 5e-8,- "input_cached_tokens": 5e-9,- "output": 4e-7,- "input_cache_read": 5e-9,- "output_reasoning_tokens": 4e-7,- "output_reasoning": 4e-7+ "input": 0.05e-6,+ "input_cached_tokens": 0.005e-6,+ "output": 0.4e-6,+ "input_cache_read": 0.005e-6,+ "output_reasoning_tokens": 0.4e-6,+ "output_reasoning": 0.4e-6,+ "cache_read_input_tokens": 0.005e-6,+ "reasoning_tokens": 0.4e-6+ }+ },+ {+ "id": "736cfc37-cb89-4981-aa0e-ffc352330dc0",+ "name": "Flex",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.025e-6,+ "input_cached_tokens": 0.0025e-6,+ "output": 0.2e-6,+ "input_cache_read": 0.0025e-6,+ "output_reasoning_tokens": 0.2e-6,+ "output_reasoning": 0.2e-6,+ "cache_read_input_tokens": 0.0025e-6,+ "reasoning_tokens": 0.2e-6}}]@@ -3179,7 +3897,7 @@"modelName": "gpt-5.1","matchPattern": "(?i)^(openai\/)?(gpt-5.1)$","createdAt": "2025-11-14T08:57:23.481Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -3194,12 +3912,62 @@"priority": 0,"conditions": [],"prices": {- "input": 0.00000125,- "input_cached_tokens": 1.25e-7,- "output": 0.00001,- "input_cache_read": 1.25e-7,- "output_reasoning_tokens": 0.00001,- "output_reasoning": 0.00001+ "input": 1.25e-6,+ "input_cached_tokens": 0.125e-6,+ "output": 10e-6,+ "input_cache_read": 0.125e-6,+ "output_reasoning_tokens": 10e-6,+ "output_reasoning": 10e-6,+ "cache_read_input_tokens": 0.125e-6,+ "reasoning_tokens": 10e-6+ }+ },+ {+ "id": "78484894-8814-4eff-b250-06368c7c1fbb",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 2.5e-6,+ "input_cached_tokens": 0.25e-6,+ "output": 20e-6,+ "input_cache_read": 0.25e-6,+ "output_reasoning_tokens": 20e-6,+ "output_reasoning": 20e-6,+ "cache_read_input_tokens": 0.25e-6,+ "reasoning_tokens": 20e-6+ }+ },+ {+ "id": "34bf46b6-995a-4e2e-a839-430c1da110e8",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.625e-6,+ "input_cached_tokens": 0.0625e-6,+ "output": 5e-6,+ "input_cache_read": 0.0625e-6,+ "output_reasoning_tokens": 5e-6,+ "output_reasoning": 5e-6,+ "cache_read_input_tokens": 0.0625e-6,+ "reasoning_tokens": 5e-6}}]@@ -3209,7 +3977,7 @@"modelName": "gpt-5.1-2025-11-13","matchPattern": "(?i)^(openai\/)?(gpt-5.1-2025-11-13)$","createdAt": "2025-11-14T08:57:23.481Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -3224,12 +3992,62 @@"priority": 0,"conditions": [],"prices": {- "input": 0.00000125,- "input_cached_tokens": 1.25e-7,- "output": 0.00001,- "input_cache_read": 1.25e-7,- "output_reasoning_tokens": 0.00001,- "output_reasoning": 0.00001+ "input": 1.25e-6,+ "input_cached_tokens": 0.125e-6,+ "output": 10e-6,+ "input_cache_read": 0.125e-6,+ "output_reasoning_tokens": 10e-6,+ "output_reasoning": 10e-6,+ "cache_read_input_tokens": 0.125e-6,+ "reasoning_tokens": 10e-6+ }+ },+ {+ "id": "aa398ff9-69fa-4f7a-b18a-6085e82ee20b",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 2.5e-6,+ "input_cached_tokens": 0.25e-6,+ "output": 20e-6,+ "input_cache_read": 0.25e-6,+ "output_reasoning_tokens": 20e-6,+ "output_reasoning": 20e-6,+ "cache_read_input_tokens": 0.25e-6,+ "reasoning_tokens": 20e-6+ }+ },+ {+ "id": "b92ba7b5-513c-4f78-8e2b-f98adceee0fd",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.625e-6,+ "input_cached_tokens": 0.0625e-6,+ "output": 5e-6,+ "input_cache_read": 0.0625e-6,+ "output_reasoning_tokens": 5e-6,+ "output_reasoning": 5e-6,+ "cache_read_input_tokens": 0.0625e-6,+ "reasoning_tokens": 5e-6}}]@@ -3363,7 +4181,7 @@"modelName": "claude-opus-4-8","matchPattern": "(?i)^((anthropic\/)?claude-opus-4-8|(eu\\.|us\\.|apac\\.|au\\.|jp\\.|global\\.)?anthropic\\.claude-opus-4-8(-v1(:0)?)?)$","createdAt": "2026-05-28T00:00:00.000Z",- "updatedAt": "2026-07-28T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": null,"tokenizerId": "claude","pricingTiers": [@@ -3386,6 +4204,33 @@"input_cached_tokens": 0.5e-6,"input_cache_read": 0.5e-6}+ },+ {+ "id": "d157528f-9083-43be-ba65-2d281a8891db",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "speed",+ "operator": "in",+ "values": ["fast"]+ }+ ],+ "prices": {+ "input": 10e-6,+ "input_tokens": 10e-6,+ "output": 50e-6,+ "output_tokens": 50e-6,+ "cache_creation_input_tokens": 12.5e-6,+ "input_cache_creation": 12.5e-6,+ "input_cache_creation_5m": 12.5e-6,+ "input_cache_creation_1h": 20e-6,+ "cache_read_input_tokens": 1e-6,+ "input_cached_tokens": 1e-6,+ "input_cache_read": 1e-6+ }}]},@@ -3487,7 +4332,7 @@"modelName": "claude-opus-5","matchPattern": "(?i)^((anthropic\\/)?claude-opus-5|(eu\\.|us\\.|apac\\.|au\\.|jp\\.|global\\.)?anthropic\\.claude-opus-5(-v1(:0)?)?)$","createdAt": "2026-07-25T00:00:00.000Z",- "updatedAt": "2026-07-28T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": null,"tokenizerId": "claude","pricingTiers": [@@ -3510,6 +4355,33 @@"input_cached_tokens": 0.5e-6,"input_cache_read": 0.5e-6}+ },+ {+ "id": "c4a83d8c-31d6-41c8-a836-3837999083d2",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "speed",+ "operator": "in",+ "values": ["fast"]+ }+ ],+ "prices": {+ "input": 10e-6,+ "input_tokens": 10e-6,+ "output": 50e-6,+ "output_tokens": 50e-6,+ "cache_creation_input_tokens": 12.5e-6,+ "input_cache_creation": 12.5e-6,+ "input_cache_creation_5m": 12.5e-6,+ "input_cache_creation_1h": 20e-6,+ "cache_read_input_tokens": 1e-6,+ "input_cached_tokens": 1e-6,+ "input_cache_read": 1e-6+ }}]},@@ -3732,7 +4604,7 @@"modelName": "gpt-5.2","matchPattern": "(?i)^(openai\/)?(gpt-5.2)$","createdAt": "2025-12-12T09:00:06.513Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -3752,17 +4624,67 @@"input_cache_read": 0.175e-6,"output": 14e-6,"output_reasoning_tokens": 14e-6,- "output_reasoning": 14e-6+ "output_reasoning": 14e-6,+ "cache_read_input_tokens": 0.175e-6,+ "reasoning_tokens": 14e-6}- }- ]- },- {- "id": "cmj2muxg6000104kzd2tc8953",- "modelName": "gpt-5.2-2025-12-11",+ },+ {+ "id": "b9f9277b-3d3e-4c75-9559-3ae60a8e221a",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 3.5e-6,+ "input_cached_tokens": 0.35e-6,+ "input_cache_read": 0.35e-6,+ "output": 28e-6,+ "output_reasoning_tokens": 28e-6,+ "output_reasoning": 28e-6,+ "cache_read_input_tokens": 0.35e-6,+ "reasoning_tokens": 28e-6+ }+ },+ {+ "id": "74ae8425-2dbb-4879-a0da-c76a3a722c14",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.875e-6,+ "input_cached_tokens": 0.0875e-6,+ "input_cache_read": 0.0875e-6,+ "output": 7e-6,+ "output_reasoning_tokens": 7e-6,+ "output_reasoning": 7e-6,+ "cache_read_input_tokens": 0.0875e-6,+ "reasoning_tokens": 7e-6+ }+ }+ ]+ },+ {+ "id": "cmj2muxg6000104kzd2tc8953",+ "modelName": "gpt-5.2-2025-12-11","matchPattern": "(?i)^(openai\/)?(gpt-5.2-2025-12-11)$","createdAt": "2025-12-12T09:00:06.513Z",- "updatedAt": "2025-12-12T15:00:06.513Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -3782,7 +4704,57 @@"input_cache_read": 0.175e-6,"output": 14e-6,"output_reasoning_tokens": 14e-6,- "output_reasoning": 14e-6+ "output_reasoning": 14e-6,+ "cache_read_input_tokens": 0.175e-6,+ "reasoning_tokens": 14e-6+ }+ },+ {+ "id": "7616bc9c-1606-4318-a79b-b0a9e4746477",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 3.5e-6,+ "input_cached_tokens": 0.35e-6,+ "input_cache_read": 0.35e-6,+ "output": 28e-6,+ "output_reasoning_tokens": 28e-6,+ "output_reasoning": 28e-6,+ "cache_read_input_tokens": 0.35e-6,+ "reasoning_tokens": 28e-6+ }+ },+ {+ "id": "c6660696-8b14-4a83-8eb5-f1c68840a832",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.875e-6,+ "input_cached_tokens": 0.0875e-6,+ "input_cache_read": 0.0875e-6,+ "output": 7e-6,+ "output_reasoning_tokens": 7e-6,+ "output_reasoning": 7e-6,+ "cache_read_input_tokens": 0.0875e-6,+ "reasoning_tokens": 7e-6}}]@@ -3878,7 +4850,7 @@"modelName": "gpt-5.4","matchPattern": "(?i)^(openai\/)?(gpt-5.4)$","createdAt": "2026-03-05T00:00:00.000Z",- "updatedAt": "2026-08-04T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -3903,11 +4875,89 @@"reasoning_tokens": 15e-6}},+ {+ "id": "1a2e3fd0-8bbd-4b8b-95d3-1d9750c45548",+ "name": "Flex · Large context (>272K)",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ },+ {+ "usageDetailPattern": "(input)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 2.5e-6,+ "input_cached_tokens": 0.25e-6,+ "input_cache_read": 0.25e-6,+ "cache_read_input_tokens": 0.25e-6,+ "output": 11.25e-6,+ "output_reasoning_tokens": 11.25e-6,+ "output_reasoning": 11.25e-6,+ "reasoning_tokens": 11.25e-6+ }+ },+ {+ "id": "e99b8dc7-8bb2-4373-9b03-8473958c9574",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 5e-6,+ "input_cached_tokens": 0.5e-6,+ "input_cache_read": 0.5e-6,+ "cache_read_input_tokens": 0.5e-6,+ "output": 30e-6,+ "output_reasoning_tokens": 30e-6,+ "output_reasoning": 30e-6,+ "reasoning_tokens": 30e-6+ }+ },+ {+ "id": "4d8bb956-4374-430c-bfb5-66237dcbc636",+ "name": "Flex",+ "isDefault": false,+ "priority": 3,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 1.25e-6,+ "input_cached_tokens": 0.13e-6,+ "input_cache_read": 0.13e-6,+ "cache_read_input_tokens": 0.13e-6,+ "output": 7.5e-6,+ "output_reasoning_tokens": 7.5e-6,+ "output_reasoning": 7.5e-6,+ "reasoning_tokens": 7.5e-6+ }+ },{"id": "8f7d2163-197d-4bb2-a403-957c1f9f7244","name": "Large Context (>272K)","isDefault": false,- "priority": 1,+ "priority": 4,"conditions": [{"usageDetailPattern": "(input)",@@ -3934,7 +4984,7 @@"modelName": "gpt-5.4-pro","matchPattern": "(?i)^(openai\/)?(gpt-5.4-pro)$","createdAt": "2026-03-05T00:00:00.000Z",- "updatedAt": "2026-08-04T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -3956,11 +5006,59 @@"reasoning_tokens": 180e-6}},+ {+ "id": "303abe77-afd3-42d7-b4a9-fc259b2c20c0",+ "name": "Flex · Large context (>272K)",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ },+ {+ "usageDetailPattern": "(input)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 30e-6,+ "output": 135e-6,+ "output_reasoning_tokens": 135e-6,+ "output_reasoning": 135e-6,+ "reasoning_tokens": 135e-6+ }+ },+ {+ "id": "84c1acad-a05b-499d-97c9-8998b4ff0c5b",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 15e-6,+ "output": 90e-6,+ "output_reasoning_tokens": 90e-6,+ "output_reasoning": 90e-6,+ "reasoning_tokens": 90e-6+ }+ },{"id": "3c335f8e-6594-4b8b-abde-f3be0bcf0349","name": "Large Context (>272K)","isDefault": false,- "priority": 1,+ "priority": 3,"conditions": [{"usageDetailPattern": "(input)",@@ -3984,7 +5082,7 @@"modelName": "gpt-5.4-2026-03-05","matchPattern": "(?i)^(openai\/)?(gpt-5.4-2026-03-05)$","createdAt": "2026-03-05T00:00:00.000Z",- "updatedAt": "2026-08-04T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4009,11 +5107,89 @@"reasoning_tokens": 15e-6}},+ {+ "id": "e670a6f6-c046-445e-a6f3-d1c71d934051",+ "name": "Flex · Large context (>272K)",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ },+ {+ "usageDetailPattern": "(input)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 2.5e-6,+ "input_cached_tokens": 0.25e-6,+ "input_cache_read": 0.25e-6,+ "cache_read_input_tokens": 0.25e-6,+ "output": 11.25e-6,+ "output_reasoning_tokens": 11.25e-6,+ "output_reasoning": 11.25e-6,+ "reasoning_tokens": 11.25e-6+ }+ },+ {+ "id": "fb92ade7-c15c-4b2a-a5aa-b5f5eb6f8c34",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 5e-6,+ "input_cached_tokens": 0.5e-6,+ "input_cache_read": 0.5e-6,+ "cache_read_input_tokens": 0.5e-6,+ "output": 30e-6,+ "output_reasoning_tokens": 30e-6,+ "output_reasoning": 30e-6,+ "reasoning_tokens": 30e-6+ }+ },+ {+ "id": "efb0cdd3-3a13-49dc-9b4e-29fbc9fecdcf",+ "name": "Flex",+ "isDefault": false,+ "priority": 3,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 1.25e-6,+ "input_cached_tokens": 0.13e-6,+ "input_cache_read": 0.13e-6,+ "cache_read_input_tokens": 0.13e-6,+ "output": 7.5e-6,+ "output_reasoning_tokens": 7.5e-6,+ "output_reasoning": 7.5e-6,+ "reasoning_tokens": 7.5e-6+ }+ },{"id": "596cb5e9-ba20-4fcf-aa07-fa8c557b643c","name": "Large Context (>272K)","isDefault": false,- "priority": 1,+ "priority": 4,"conditions": [{"usageDetailPattern": "(input)",@@ -4040,7 +5216,7 @@"modelName": "gpt-5.4-pro-2026-03-05","matchPattern": "(?i)^(openai\/)?(gpt-5.4-pro-2026-03-05)$","createdAt": "2026-03-05T00:00:00.000Z",- "updatedAt": "2026-08-04T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4062,11 +5238,59 @@"reasoning_tokens": 180e-6}},+ {+ "id": "af0f392c-1286-4cac-a7be-6b85d9c5449e",+ "name": "Flex · Large context (>272K)",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ },+ {+ "usageDetailPattern": "(input)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 30e-6,+ "output": 135e-6,+ "output_reasoning_tokens": 135e-6,+ "output_reasoning": 135e-6,+ "reasoning_tokens": 135e-6+ }+ },+ {+ "id": "e603b77a-ce1f-4306-b345-322c9b37cc44",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 15e-6,+ "output": 90e-6,+ "output_reasoning_tokens": 90e-6,+ "output_reasoning": 90e-6,+ "reasoning_tokens": 90e-6+ }+ },{"id": "a5aec834-b2a6-4d42-addc-c9c424034cb4","name": "Large Context (>272K)","isDefault": false,- "priority": 1,+ "priority": 3,"conditions": [{"usageDetailPattern": "(input)",@@ -4090,7 +5314,7 @@"modelName": "gpt-5.4-mini","matchPattern": "(?i)^(openai\\/)?(gpt-5.4-mini)$","createdAt": "2026-03-18T00:00:00.000Z",- "updatedAt": "2026-03-18T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4108,7 +5332,50 @@"input": 0.75e-6,"input_cached_tokens": 0.075e-6,"input_cache_read": 0.075e-6,- "output": 4.5e-6+ "output": 4.5e-6,+ "cache_read_input_tokens": 0.075e-6+ }+ },+ {+ "id": "33d3b03a-fa64-46ce-b077-fd245bc7997d",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 1.5e-6,+ "input_cached_tokens": 0.15e-6,+ "input_cache_read": 0.15e-6,+ "output": 9e-6,+ "cache_read_input_tokens": 0.15e-6+ }+ },+ {+ "id": "386cf9dc-77cb-4d09-9cb1-e5eae3e69abd",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.375e-6,+ "input_cached_tokens": 0.0375e-6,+ "input_cache_read": 0.0375e-6,+ "output": 2.25e-6,+ "cache_read_input_tokens": 0.0375e-6}}]@@ -4118,7 +5385,7 @@"modelName": "gpt-5.4-mini-2026-03-17","matchPattern": "(?i)^(openai\\/)?(gpt-5.4-mini-2026-03-17)$","createdAt": "2026-03-18T00:00:00.000Z",- "updatedAt": "2026-03-18T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4127,16 +5394,59 @@"tokenizerId": "openai","pricingTiers": [{- "id": "3d8027df-7491-442e-9770-1363ab2452fc_tier_default",- "name": "Standard",- "isDefault": true,- "priority": 0,- "conditions": [],+ "id": "3d8027df-7491-442e-9770-1363ab2452fc_tier_default",+ "name": "Standard",+ "isDefault": true,+ "priority": 0,+ "conditions": [],+ "prices": {+ "input": 0.75e-6,+ "input_cached_tokens": 0.075e-6,+ "input_cache_read": 0.075e-6,+ "output": 4.5e-6,+ "cache_read_input_tokens": 0.075e-6+ }+ },+ {+ "id": "a4470b39-40df-45a2-a77b-b1e5e4cea7e0",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 1.5e-6,+ "input_cached_tokens": 0.15e-6,+ "input_cache_read": 0.15e-6,+ "output": 9e-6,+ "cache_read_input_tokens": 0.15e-6+ }+ },+ {+ "id": "8c294d78-ef7d-469c-9669-e2a515ca028e",+ "name": "Flex",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],"prices": {- "input": 0.75e-6,- "input_cached_tokens": 0.075e-6,- "input_cache_read": 0.075e-6,- "output": 4.5e-6+ "input": 0.375e-6,+ "input_cached_tokens": 0.0375e-6,+ "input_cache_read": 0.0375e-6,+ "output": 2.25e-6,+ "cache_read_input_tokens": 0.0375e-6}}]@@ -4146,7 +5456,7 @@"modelName": "gpt-5.4-nano","matchPattern": "(?i)^(openai\\/)?(gpt-5.4-nano)$","createdAt": "2026-03-18T00:00:00.000Z",- "updatedAt": "2026-03-18T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4164,7 +5474,29 @@"input": 0.2e-6,"input_cached_tokens": 0.02e-6,"input_cache_read": 0.02e-6,- "output": 1.25e-6+ "output": 1.25e-6,+ "cache_read_input_tokens": 0.02e-6+ }+ },+ {+ "id": "12454d21-744a-4c31-b0a3-54d1725015da",+ "name": "Flex",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.1e-6,+ "input_cached_tokens": 0.01e-6,+ "input_cache_read": 0.01e-6,+ "output": 0.625e-6,+ "cache_read_input_tokens": 0.01e-6}}]@@ -4174,7 +5506,7 @@"modelName": "gpt-5.4-nano-2026-03-17","matchPattern": "(?i)^(openai\\/)?(gpt-5.4-nano-2026-03-17)$","createdAt": "2026-03-18T00:00:00.000Z",- "updatedAt": "2026-03-18T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4192,7 +5524,29 @@"input": 0.2e-6,"input_cached_tokens": 0.02e-6,"input_cache_read": 0.02e-6,- "output": 1.25e-6+ "output": 1.25e-6,+ "cache_read_input_tokens": 0.02e-6+ }+ },+ {+ "id": "e5e91108-dbc1-44a7-930b-feb35141ed70",+ "name": "Flex",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.1e-6,+ "input_cached_tokens": 0.01e-6,+ "input_cache_read": 0.01e-6,+ "output": 0.625e-6,+ "cache_read_input_tokens": 0.01e-6}}]@@ -4202,7 +5556,7 @@"modelName": "gpt-5.5-2026-04-23","matchPattern": "(?i)^(openai\/)?(gpt-5.5(-2026-04-23)?)$","createdAt": "2026-04-25T00:00:00.000Z",- "updatedAt": "2026-08-18T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4227,11 +5581,41 @@"reasoning_tokens": 30e-6}},+ {+ "id": "67ba2a0f-c796-4d93-bbb6-cf0338718150",+ "name": "Flex · Large context (>272K)",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ },+ {+ "usageDetailPattern": "(input)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 5e-6,+ "input_cached_tokens": 0.5e-6,+ "input_cache_read": 0.5e-6,+ "cache_read_input_tokens": 0.5e-6,+ "output": 22.5e-6,+ "output_reasoning_tokens": 22.5e-6,+ "output_reasoning": 22.5e-6,+ "reasoning_tokens": 22.5e-6+ }+ },{"id": "af5ffdbd-411b-4215-ae86-bfc0d9c9817e","name": "Fast mode","isDefault": false,- "priority": 1,+ "priority": 2,"conditions": [{"source": "model_parameters",@@ -4251,11 +5635,35 @@"reasoning_tokens": 75e-6}},+ {+ "id": "93bdcf85-5a30-4cfa-8144-25d4f7af82ec",+ "name": "Flex",+ "isDefault": false,+ "priority": 3,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 2.5e-6,+ "input_cached_tokens": 0.25e-6,+ "input_cache_read": 0.25e-6,+ "cache_read_input_tokens": 0.25e-6,+ "output": 15e-6,+ "output_reasoning_tokens": 15e-6,+ "output_reasoning": 15e-6,+ "reasoning_tokens": 15e-6+ }+ },{"id": "a3ad657c-fff1-44dd-a51b-20d9845a112a","name": "Large Context (>272K)","isDefault": false,- "priority": 2,+ "priority": 4,"conditions": [{"usageDetailPattern": "(input)",@@ -4282,7 +5690,7 @@"modelName": "gpt-5.5-pro-2026-04-23","matchPattern": "(?i)^(openai\/)?(gpt-5.5-pro(-2026-04-23)?)$","createdAt": "2026-04-25T00:00:00.000Z",- "updatedAt": "2026-08-04T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4304,11 +5712,32 @@"reasoning_tokens": 180e-6}},+ {+ "id": "921195dc-7b97-44c7-a12e-0dd4acafe335",+ "name": "Flex",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 15e-6,+ "output": 90e-6,+ "output_reasoning_tokens": 90e-6,+ "output_reasoning": 90e-6,+ "reasoning_tokens": 90e-6+ }+ },{"id": "56beb248-e8af-4d42-a861-1f921e81cfb2","name": "Large Context (>272K)","isDefault": false,- "priority": 1,+ "priority": 2,"conditions": [{"usageDetailPattern": "(input)",@@ -4332,7 +5761,7 @@"modelName": "gpt-5.6-sol","matchPattern": "(?i)^(openai/)?(gpt-5.6-sol)$","createdAt": "2026-07-09T00:00:00.000Z",- "updatedAt": "2026-08-18T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4391,11 +5820,43 @@"reasoning_tokens": 90e-6}},+ {+ "id": "53c9716e-6693-4ad4-af66-a00091b91d80",+ "name": "Flex · Large context (>272K)",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ },+ {+ "usageDetailPattern": "(input|cache_write)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 5e-6,+ "input_cached_tokens": 0.5e-6,+ "input_cache_read": 0.5e-6,+ "cache_read_input_tokens": 0.5e-6,+ "input_cache_creation": 6.25e-6,+ "cache_write_tokens": 6.25e-6,+ "output": 22.5e-6,+ "output_reasoning_tokens": 22.5e-6,+ "output_reasoning": 22.5e-6,+ "reasoning_tokens": 22.5e-6+ }+ },{"id": "26ef382b-df27-4a55-ad8c-dedea29d76ee","name": "Fast mode","isDefault": false,- "priority": 2,+ "priority": 3,"conditions": [{"source": "model_parameters",@@ -4417,11 +5878,37 @@"reasoning_tokens": 60e-6}},+ {+ "id": "4735e32d-5ff7-4fc9-ad62-8c9579dc5f2a",+ "name": "Flex",+ "isDefault": false,+ "priority": 4,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 2.5e-6,+ "input_cached_tokens": 0.25e-6,+ "input_cache_read": 0.25e-6,+ "cache_read_input_tokens": 0.25e-6,+ "input_cache_creation": 3.125e-6,+ "cache_write_tokens": 3.125e-6,+ "output": 15e-6,+ "output_reasoning_tokens": 15e-6,+ "output_reasoning": 15e-6,+ "reasoning_tokens": 15e-6+ }+ },{"id": "1a2b3c4d-5e6f-7a8b-9c0d-1e2f3a4b5c6d","name": "Large Context (>272K)","isDefault": false,- "priority": 3,+ "priority": 5,"conditions": [{"usageDetailPattern": "(input|cache_write)",@@ -4450,7 +5937,7 @@"modelName": "gpt-5.6-terra","matchPattern": "(?i)^(openai/)?(gpt-5.6-terra)$","createdAt": "2026-07-09T00:00:00.000Z",- "updatedAt": "2026-08-18T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4509,11 +5996,43 @@"reasoning_tokens": 36e-6}},+ {+ "id": "8a5adb80-6e59-4a56-a69f-9eda3d8ebaae",+ "name": "Flex · Large context (>272K)",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ },+ {+ "usageDetailPattern": "(input|cache_write)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 2e-6,+ "input_cached_tokens": 0.2e-6,+ "input_cache_read": 0.2e-6,+ "cache_read_input_tokens": 0.2e-6,+ "input_cache_creation": 2.5e-6,+ "cache_write_tokens": 2.5e-6,+ "output": 9e-6,+ "output_reasoning_tokens": 9e-6,+ "output_reasoning": 9e-6,+ "reasoning_tokens": 9e-6+ }+ },{"id": "9ff3bb71-f328-4cee-81b8-6b767871d820","name": "Fast mode","isDefault": false,- "priority": 2,+ "priority": 3,"conditions": [{"source": "model_parameters",@@ -4535,11 +6054,37 @@"reasoning_tokens": 24e-6}},+ {+ "id": "489d530a-bcfd-4667-a4a3-508ff3555cf7",+ "name": "Flex",+ "isDefault": false,+ "priority": 4,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 1e-6,+ "input_cached_tokens": 0.1e-6,+ "input_cache_read": 0.1e-6,+ "cache_read_input_tokens": 0.1e-6,+ "input_cache_creation": 1.25e-6,+ "cache_write_tokens": 1.25e-6,+ "output": 6e-6,+ "output_reasoning_tokens": 6e-6,+ "output_reasoning": 6e-6,+ "reasoning_tokens": 6e-6+ }+ },{"id": "2b3c4d5e-6f7a-8b9c-0d1e-2f3a4b5c6d7e","name": "Large Context (>272K)","isDefault": false,- "priority": 3,+ "priority": 5,"conditions": [{"usageDetailPattern": "(input|cache_write)",@@ -4568,7 +6113,7 @@"modelName": "gpt-5.6-luna","matchPattern": "(?i)^(openai/)?(gpt-5.6-luna)$","createdAt": "2026-07-09T00:00:00.000Z",- "updatedAt": "2026-08-18T00:00:00.000Z",+ "updatedAt": "2026-08-20T00:00:00.000Z","tokenizerConfig": {"tokensPerName": 1,"tokenizerModel": "gpt-4",@@ -4627,11 +6172,43 @@"reasoning_tokens": 3.6e-6}},+ {+ "id": "dde1aa4b-c9ba-46ad-803a-561411629f77",+ "name": "Flex · Large context (>272K)",+ "isDefault": false,+ "priority": 2,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ },+ {+ "usageDetailPattern": "(input|cache_write)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 0.2e-6,+ "input_cached_tokens": 0.02e-6,+ "input_cache_read": 0.02e-6,+ "cache_read_input_tokens": 0.02e-6,+ "input_cache_creation": 0.25e-6,+ "cache_write_tokens": 0.25e-6,+ "output": 0.9e-6,+ "output_reasoning_tokens": 0.9e-6,+ "output_reasoning": 0.9e-6,+ "reasoning_tokens": 0.9e-6+ }+ },{"id": "079cdd36-4c31-4df3-9e4b-38074becb0b4","name": "Fast mode","isDefault": false,- "priority": 2,+ "priority": 3,"conditions": [{"source": "model_parameters",@@ -4653,11 +6230,37 @@"reasoning_tokens": 2.4e-6}},+ {+ "id": "2a50074e-4a63-4a3f-b161-fd8811b3f655",+ "name": "Flex",+ "isDefault": false,+ "priority": 4,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.1e-6,+ "input_cached_tokens": 0.01e-6,+ "input_cache_read": 0.01e-6,+ "cache_read_input_tokens": 0.01e-6,+ "input_cache_creation": 0.125e-6,+ "cache_write_tokens": 0.125e-6,+ "output": 0.6e-6,+ "output_reasoning_tokens": 0.6e-6,+ "output_reasoning": 0.6e-6,+ "reasoning_tokens": 0.6e-6+ }+ },{"id": "3c4d5e6f-7a8b-9c0d-1e2f-3a4b5c6d7e8f","name": "Large Context (>272K)","isDefault": false,- "priority": 3,+ "priority": 5,"conditions": [{"usageDetailPattern": "(input|cache_write)",