πͺ’ 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 Claude Opus 5.5 and GPT-6 Sol/Luna default pricing (#17802)
langfuse-botb2c4d3f5c3d3
diff --git a/.agents/skills/add-model-price/references/model-audit-memory.md b/.agents/skills/add-model-price/references/model-audit-memory.mdindex 9a50db7c22..e14f1757a7 100644--- a/.agents/skills/add-model-price/references/model-audit-memory.md+++ b/.agents/skills/add-model-price/references/model-audit-memory.md@@ -12,11 +12,25 @@ audit date.## Latest useful snapshot-**Audit date:** 2026-09-14+**Audit date:** 2026-09-22All prices listed as `$X / MTok` (per million tokens). Per-token JSON values: divide by 1,000,000.-The 2026-09-14 run re-fetched the full Anthropic pricing table (plus the+The 2026-09-22 run re-fetched the full Anthropic pricing page, the Anthropic+models-overview comparison table, the OpenAI aggregate Standard/Batch/Flex/Fast-mode+pricing tables plus the full model catalog, and both Gemini pricing pages (3.x and+2.5 families) plus the Gemini models catalog, with a follow-up verbatim-quote fetch+for `gemini-3.5-flash-lite`/`gemini-3.1-flash-lite`/`gemini-3.1-pro-preview`/+`gemini-3-flash-preview`. This run found two new model families and added all+three of their members: **Claude Opus 5.5** (`claude-opus-5-5`, Anthropic's new+recommended default, priced at half of Claude Opus 5 with a non-standard 0.05x+cache-read multiplier) and **GPT-6 Sol** / **GPT-6 Luna** (`gpt-6-sol`,+`gpt-6-luna`, filling out the GPT-6 family below `gpt-6-astra` with the same+six-tier Standard/Fast-mode/Flex/Large-Context shape). Every other price checked+this run β including every existing Anthropic, OpenAI, and Gemini row below β+matched verbatim; see `provider-sources-and-price-keys.md`'s "September 22 2026+audit" entry for full detail and the new models' own dedicated entries for exact+figures and matchPatterns. The 2026-09-14 run before it re-fetched the full Anthropic pricing table (plus themodels-overview table), the full OpenAI standard/Fast-mode/Flex pricing tables plusthe full model catalog page, both Gemini pricing pages (2.5-family and 3.x-family)plus the Gemini models catalog, and a dedicated re-fetch of `gpt-5-chat-latest`'s own@@ -46,7 +60,9 @@ to have siblings `gpt-5.5-cyber`/`gpt-5.4-cyber`, see provider-sources-and-price| Provider | Model / pricing entry | Pricing checked | Price confirmed | Tiering checked | Tiering correct | Change | Official source(s) | Comments || -------- | --------------------- | --------------- | --------------- | --------------- | --------------- | ------ | ------------------- | -------- |-| OpenAI | gpt-6-astra | Input $10/MTok, Cached $1/MTok, Cache write $12.50/MTok, Output $50/MTok | Yes | Large Context (>272K) 2x/2x/2x/1.5x; Fast mode 2x base; Flex 0.5x base | Yes | None | https://developers.openai.com/api/docs/pricing https://developers.openai.com/api/docs/models/gpt-6-astra | Added 2026-09-03, re-confirmed unchanged 2026-09-10 and 2026-09-14. Six-tier gpt-5.6-sol-style key set. |+| OpenAI | gpt-6-astra | Input $10/MTok, Cached $1/MTok, Cache write $12.50/MTok, Output $50/MTok | Yes | Large Context (>272K) 2x/2x/2x/1.5x; Fast mode 2x base; Flex 0.5x base | Yes | None | https://developers.openai.com/api/docs/pricing https://developers.openai.com/api/docs/models/gpt-6-astra | Added 2026-09-03, re-confirmed unchanged 2026-09-10, 2026-09-14, and 2026-09-22. Six-tier gpt-5.6-sol-style key set. |+| OpenAI | gpt-6-sol | Input $2/MTok, Cached $0.20/MTok, Cache write $2.50/MTok, Output $10/MTok | Yes | Large Context (>272K) 2x/2x/2x/1.5x; Fast mode 2x applicable tier; Flex 0.5x applicable tier | Yes | Added | https://developers.openai.com/api/docs/pricing https://developers.openai.com/api/docs/models/gpt-6-sol | Added 2026-09-22. Six-tier gpt-6-astra-style key set; same 1,050,000-token context window and 272K large-context threshold as gpt-6-astra. |+| OpenAI | gpt-6-luna | Input $0.10/MTok, Cached $0.01/MTok, Cache write $0.125/MTok, Output $0.50/MTok | Yes | Large Context (>272K) 2x/2x/2x/1.5x; Fast mode 2x applicable tier; Flex 0.5x applicable tier | Yes | Added | https://developers.openai.com/api/docs/pricing https://developers.openai.com/api/docs/models/gpt-6-luna | Added 2026-09-22. Six-tier gpt-6-astra-style key set; same context window and threshold as gpt-6-astra/gpt-6-sol. || Provider | Model / pricing entry | Pricing checked | Price confirmed | Tiering checked | Tiering correct | Change | Official source(s) | Comments || -------- | --------------------- | --------------- | --------------- | --------------- | --------------- | ------ | ------------------ | -------- |@@ -54,6 +70,7 @@ to have siblings `gpt-5.5-cyber`/`gpt-5.4-cyber`, see provider-sources-and-price| Anthropic | claude-mythos-5-1 | Same as claude-fable-5-1 | Yes | Flat 1M context; no Fast mode | Yes | Added | https://platform.claude.com/docs/en/about-claude/pricing https://platform.claude.com/docs/en/models/mythos-5-1/overview | Limited availability (Project Glasswing), mirrors claude-mythos-5's simpler API-only matchPattern (no Bedrock/GCP variants). || Anthropic | claude-fable-5 | Input $10/MTok, Output $50/MTok, 5m $12.50/MTok, 1h $20/MTok, read $1/MTok | Yes | Flat 1M context | Yes | None | https://platform.claude.com/docs/en/about-claude/pricing | Re-confirmed unchanged; now listed as "legacy (still available)" behind Fable 5.1 but pricing is the same. || Anthropic | claude-mythos-5 | Same as claude-fable-5 | Yes | Flat 1M context | Yes | None | https://platform.claude.com/docs/en/about-claude/pricing | Re-confirmed unchanged. |+| Anthropic | claude-opus-5-5 | Input $4/MTok, Output $20/MTok, 5m $5/MTok, 1h $8/MTok, read $0.20/MTok (0.05x, not the standard 0.1x) | Yes | Flat 1M context; Fast mode $8/$40 confirmed available | Yes | Added | https://platform.claude.com/docs/en/about-claude/pricing https://platform.claude.com/docs/en/models/overview | New model, now Anthropic's recommended default ("For long-running agentic coding and knowledge work") ahead of Opus 5. Cache-read discount is 0.05x per an explicit page footnote (a third distinct multiplier alongside the 0.025x Fable 5.1/Mythos 5.1 rate and the standard 0.1x). Fast-mode cache read/write derived by applying the documented multipliers to the $8 Fast-mode base (not the $4 Standard base): read $0.40, 5m write $10, 1h write $16. matchPattern mirrors claude-opus-5 with a `-5` suffix. || Anthropic | claude-opus-5 | Input $5/MTok, Output $25/MTok, 5m $6.25/MTok, 1h $10/MTok, read $0.50/MTok | Yes | Flat 1M context; Fast mode $10/$50 confirmed available | Yes | None | https://platform.claude.com/docs/en/about-claude/pricing | Re-confirmed unchanged. || Anthropic | claude-opus-4-8 | Same as Opus 5, Fast mode $10/$50 confirmed available | Yes | Flat 1M context | Yes | None | https://platform.claude.com/docs/en/about-claude/pricing | Re-confirmed unchanged. || Anthropic | claude-opus-4-7 | Same as Opus 5 base rate; **Fast mode NOT available** (requests with `speed:"fast"` error) | Yes | Flat 1M context; no Fast mode tier | Yes | None | https://platform.claude.com/docs/en/about-claude/pricing | Explicitly confirmed via page text this run. File already correctly has only a Standard tier (no Fast mode) β do not add one without re-checking this page. |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 3ca313821e..dc6f09bb4f 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@@ -685,6 +685,84 @@ file and `openAIModels`in July 27 2026 audit. Official sources:`https://developeDaybreak/Rosalind restricted family only if OpenAI publishes public,unauthenticated documentation confirming a model ID and price on its owndedicated page.+- **Claude Opus 5.5 (added September 22 2026)** β Anthropic released+ `claude-opus-5-5`, now the recommended default on the models-overview+ comparison table ("For long-running agentic coding and knowledge work"),+ confirmed via `https://platform.claude.com/docs/en/about-claude/pricing` and+ `https://platform.claude.com/docs/en/models/overview`. API ID / alias /+ Bedrock ID / Google Cloud ID / Microsoft Foundry ID / Claude Platform on AWS+ ID are all the dateless `claude-opus-5-5` / `anthropic.claude-opus-5-5`+ pattern (mirroring `claude-opus-5`, one more `-5` segment). Pricing: $4/MTok+ input, $20/MTok output, 5m cache write $5/MTok, 1h cache write $8/MTok β all+ half of Claude Opus 5's rate. **Cache hits are priced at 0.05x base input+ ($0.20/MTok), not the standard 0.1x multiplier** β confirmed verbatim via+ the pricing page's cache-hits footnote, which now lists three non-standard+ multipliers side by side: 0.025x for Fable 5.1/Mythos 5.1, 0.05x for Opus+ 5.5, 0.1x for every other model. Fast mode is available at $8/$40 input/output+ (`speed: "fast"`, same mechanism as Opus 5/4.8); the page's Fast-mode table+ only lists Input/Output, so β consistent with how the existing+ `claude-opus-5`/`claude-opus-4-8` Fast-mode tiers were derived β the Fast-mode+ cache read/write prices were computed by applying the documented cache+ multipliers (0.05x read, 1.25x 5m write, 2x 1h write) to the *Fast-mode* base+ input price, not the Standard base input price: $0.40/MTok read, $10/MTok 5m+ write, $16/MTok 1h write. On the flat 1M-context list (no Large Context+ tier). Batch is $2/$10 (50% of standard, per the page's Batch table) but, per+ existing precedent, no Batch tier was added to the pricing file since no+ Anthropic model has ever had one represented (Batch is a distinct API+ endpoint, not a `model_parameters` condition observable in ordinary+ ingestion usage). matchPattern:+ `(?i)^((anthropic\/)?claude-opus-5-5|(eu\.|us\.|apac\.|au\.|jp\.|global\.)?anthropic\.claude-opus-5-5(-v1(:0)?)?)$`+ β verified this does not collide with `claude-opus-5`'s pattern since both+ are fully anchored with `^...$`.+- **GPT-6 Sol / GPT-6 Luna (added September 22 2026)** β OpenAI expanded the+ GPT-6 family beyond `gpt-6-astra` with two more flagship-tier models,+ confirmed via `https://developers.openai.com/api/docs/pricing` (aggregate+ Standard/Batch/Flex/Fast-mode tables) and their dedicated model pages+ `https://developers.openai.com/api/docs/models/gpt-6-sol` and `.../gpt-6-luna`.+ Both share `gpt-6-astra`'s exact shape: 1,050,000-token context window (max+ input 922,000, max output 128,000), prompt caching and reasoning tokens+ supported, and the same >272,000-input-token Large Context threshold at 2x+ input/cache and 1.5x output. `gpt-6-sol` ("complex coding and agentic+ workflows"): standard $2/$0.20/$2.50/$10 input/cached/cache-write/output;+ large context $4/$0.40/$5.00/$15; Fast mode is 2x the applicable tier+ ($4/$0.40/$5.00/$20 standard, $8/$0.80/$10.00/$30 large context β confirmed+ against the aggregate table's Fast-mode row); Flex is 0.5x the applicable+ tier ($1.00/$0.10/$1.25/$5.00 standard, $2.00/$0.20/$2.50/$7.50 large+ context β confirmed against the aggregate table's Flex row, which is+ numerically identical to the Batch row at this multiplier, consistent with+ the `gpt-6-astra`/`gpt-5.6-sol` precedent). `gpt-6-luna` ("most efficient+ model for focused, high-volume tasks"): standard $0.10/$0.01/$0.125/$0.50;+ large context $0.20/$0.02/$0.25/$0.75; Fast mode $0.20/$0.02/$0.25/$1.00+ standard, $0.40/$0.04/$0.50/$1.50 large context; Flex $0.05/$0.005/$0.0625/$0.25+ standard, $0.10/$0.01/$0.125/$0.375 large context. Both added to the pricing+ file mirroring `gpt-6-astra`'s exact six-tier key set (Standard, Fast mode Β·+ Large context, Flex Β· Large context, Fast mode, Flex, Large Context) and to+ `openAIModels` in `types.ts` (immediately after `gpt-6-astra`, not as the+ first entry). matchPatterns: `(?i)^(openai/)?(gpt-6-sol)$` and+ `(?i)^(openai/)?(gpt-6-luna)$`. Batch pricing intentionally not added, per+ the same `gpt-6-astra`-precedent scope boundary (Batch is a distinct async+ endpoint, not an ordinary `service_tier` value Langfuse observes in+ synchronous request usage).+- **September 22 2026 audit: full re-fetch found Claude Opus 5.5 and GPT-6+ Sol/Luna as the only drift; everything else confirmed unchanged** β+ Re-fetched the full Anthropic pricing page (all sections, not just the+ model table), the Anthropic models-overview comparison table, the OpenAI+ aggregate Standard/Batch/Flex/Fast-mode pricing tables plus the full model+ catalog, and both Gemini pricing pages (3.x family and 2.5 family) plus the+ Gemini models catalog, with a follow-up verbatim-quote fetch that confirmed+ full input/output/cache pricing for `gemini-3.5-flash-lite`,+ `gemini-3.1-flash-lite`, `gemini-3.1-pro-preview`, and+ `gemini-3-flash-preview` (all unchanged). No further new general-purpose+ text/chat models were found: the Gemini models catalog's new entries this+ run (`gemini-3.8-live`, `gemini-3.8-live-extended-thinking`,+ `gemini-omni-1.1-flash`, `lyria-3.5`, more `gemini-robotics-er-2-preview`+ variants) are all Live/voice, video, music, or robotics endpoints, consistent+ with the existing modality-specific skip rule. `gpt-5.3-codex` and+ `gpt-5-chat-latest` were not independently re-fetched this run (no drift+ signal for either); their prices are carried forward from the September 2+ and September 17 2026 confirmations respectively. The Daybreak+ cyber/Rosalind restricted family and the AWS Bedrock Public Extended Access+ SKU were not re-checked this run β no new evidence, standing exclusions.Capture:diff --git a/packages/shared/src/server/llm/types.ts b/packages/shared/src/server/llm/types.tsindex 10468d1c0c..0d0756c07a 100644--- a/packages/shared/src/server/llm/types.ts+++ b/packages/shared/src/server/llm/types.ts@@ -330,6 +330,8 @@ export const openAIModels = ["gpt-4.1-nano","gpt-4.1-nano-2025-04-14","gpt-6-astra",+ "gpt-6-sol",+ "gpt-6-luna","gpt-5.6-sol","gpt-5.6-terra","gpt-5.6-luna",@@ -397,6 +399,7 @@ export const anthropicModels = ["claude-fable-5-1","claude-mythos-5","claude-mythos-5-1",+ "claude-opus-5-5","claude-opus-5","claude-haiku-4-5-20251001","claude-opus-4-8",diff --git a/worker/src/constants/default-model-prices.json b/worker/src/constants/default-model-prices.jsonindex 6746e4d847..6d62cd7bc7 100644--- a/worker/src/constants/default-model-prices.json+++ b/worker/src/constants/default-model-prices.json@@ -4517,6 +4517,64 @@}]},+ {+ "id": "008a7771-38bc-46ff-81cd-a21e8526851d",+ "modelName": "claude-opus-5-5",+ "matchPattern": "(?i)^((anthropic\\/)?claude-opus-5-5|(eu\\.|us\\.|apac\\.|au\\.|jp\\.|global\\.)?anthropic\\.claude-opus-5-5(-v1(:0)?)?)$",+ "createdAt": "2026-09-22T00:00:00.000Z",+ "updatedAt": "2026-09-22T00:00:00.000Z",+ "tokenizerConfig": null,+ "tokenizerId": "claude",+ "pricingTiers": [+ {+ "id": "008a7771-38bc-46ff-81cd-a21e8526851d_tier_default",+ "name": "Standard",+ "isDefault": true,+ "priority": 0,+ "conditions": [],+ "prices": {+ "input": 4e-6,+ "input_tokens": 4e-6,+ "output": 20e-6,+ "output_tokens": 20e-6,+ "cache_creation_input_tokens": 5e-6,+ "input_cache_creation": 5e-6,+ "input_cache_creation_5m": 5e-6,+ "input_cache_creation_1h": 8e-6,+ "cache_read_input_tokens": 0.2e-6,+ "input_cached_tokens": 0.2e-6,+ "input_cache_read": 0.2e-6+ }+ },+ {+ "id": "c2993afd-fc5b-44d3-8b11-60cf6e6f239a",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "speed",+ "operator": "in",+ "values": ["fast"]+ }+ ],+ "prices": {+ "input": 8e-6,+ "input_tokens": 8e-6,+ "output": 40e-6,+ "output_tokens": 40e-6,+ "cache_creation_input_tokens": 10e-6,+ "input_cache_creation": 10e-6,+ "input_cache_creation_5m": 10e-6,+ "input_cache_creation_1h": 16e-6,+ "cache_read_input_tokens": 0.4e-6,+ "input_cached_tokens": 0.4e-6,+ "input_cache_read": 0.4e-6+ }+ }+ ]+ },{"id": "cmig1hb7i000104l72qrzgc6h","modelName": "gemini-2.5-pro",@@ -6809,6 +6867,358 @@}]},+ {+ "id": "2a17ca9a-5cfd-47ca-b4b5-b04978ff77bb",+ "modelName": "gpt-6-sol",+ "matchPattern": "(?i)^(openai/)?(gpt-6-sol)$",+ "createdAt": "2026-09-22T00:00:00.000Z",+ "updatedAt": "2026-09-22T00:00:00.000Z",+ "tokenizerConfig": {+ "tokensPerName": 1,+ "tokenizerModel": "gpt-4",+ "tokensPerMessage": 3+ },+ "tokenizerId": "openai",+ "pricingTiers": [+ {+ "id": "2a17ca9a-5cfd-47ca-b4b5-b04978ff77bb_tier_default",+ "name": "Standard",+ "isDefault": true,+ "priority": 0,+ "conditions": [],+ "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": 10e-6,+ "output_reasoning_tokens": 10e-6,+ "output_reasoning": 10e-6,+ "reasoning_tokens": 10e-6+ }+ },+ {+ "id": "1317e902-6b79-47ad-8dc5-6323f9e5836c",+ "name": "Fast mode Β· Large context (>272K)",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ },+ {+ "usageDetailPattern": "(input|cache_write)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 8e-6,+ "input_cached_tokens": 0.8e-6,+ "input_cache_read": 0.8e-6,+ "cache_read_input_tokens": 0.8e-6,+ "input_cache_creation": 10e-6,+ "cache_write_tokens": 10e-6,+ "output": 30e-6,+ "output_reasoning_tokens": 30e-6,+ "output_reasoning": 30e-6,+ "reasoning_tokens": 30e-6+ }+ },+ {+ "id": "e60a9e54-b790-4d8a-be58-f21bd5e5f465",+ "name": "Flex Β· 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": 7.5e-6,+ "output_reasoning_tokens": 7.5e-6,+ "output_reasoning": 7.5e-6,+ "reasoning_tokens": 7.5e-6+ }+ },+ {+ "id": "c287d21e-9892-4fe1-a47d-09f28c8dcb91",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 3,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "prices": {+ "input": 4e-6,+ "input_cached_tokens": 0.4e-6,+ "input_cache_read": 0.4e-6,+ "cache_read_input_tokens": 0.4e-6,+ "input_cache_creation": 5e-6,+ "cache_write_tokens": 5e-6,+ "output": 20e-6,+ "output_reasoning_tokens": 20e-6,+ "output_reasoning": 20e-6,+ "reasoning_tokens": 20e-6+ }+ },+ {+ "id": "c58d07a5-777e-4328-b6b5-9eeec095ceb1",+ "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": 5e-6,+ "output_reasoning_tokens": 5e-6,+ "output_reasoning": 5e-6,+ "reasoning_tokens": 5e-6+ }+ },+ {+ "id": "f6a7a58d-cc60-462d-8740-2334753b4c51",+ "name": "Large Context (>272K)",+ "isDefault": false,+ "priority": 5,+ "conditions": [+ {+ "usageDetailPattern": "(input|cache_write)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 4e-6,+ "input_cached_tokens": 0.4e-6,+ "input_cache_read": 0.4e-6,+ "cache_read_input_tokens": 0.4e-6,+ "input_cache_creation": 5e-6,+ "cache_write_tokens": 5e-6,+ "output": 15e-6,+ "output_reasoning_tokens": 15e-6,+ "output_reasoning": 15e-6,+ "reasoning_tokens": 15e-6+ }+ }+ ]+ },+ {+ "id": "8edc337e-ba50-4811-bc51-e9ad4253826c",+ "modelName": "gpt-6-luna",+ "matchPattern": "(?i)^(openai/)?(gpt-6-luna)$",+ "createdAt": "2026-09-22T00:00:00.000Z",+ "updatedAt": "2026-09-22T00:00:00.000Z",+ "tokenizerConfig": {+ "tokensPerName": 1,+ "tokenizerModel": "gpt-4",+ "tokensPerMessage": 3+ },+ "tokenizerId": "openai",+ "pricingTiers": [+ {+ "id": "8edc337e-ba50-4811-bc51-e9ad4253826c_tier_default",+ "name": "Standard",+ "isDefault": true,+ "priority": 0,+ "conditions": [],+ "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": 0.5e-6,+ "output_reasoning_tokens": 0.5e-6,+ "output_reasoning": 0.5e-6,+ "reasoning_tokens": 0.5e-6+ }+ },+ {+ "id": "b4525852-c687-4d0b-bf4c-f19150fbb3ab",+ "name": "Fast mode Β· Large context (>272K)",+ "isDefault": false,+ "priority": 1,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ },+ {+ "usageDetailPattern": "(input|cache_write)",+ "operator": "gt",+ "value": 272000,+ "caseSensitive": false+ }+ ],+ "prices": {+ "input": 0.4e-6,+ "input_cached_tokens": 0.04e-6,+ "input_cache_read": 0.04e-6,+ "cache_read_input_tokens": 0.04e-6,+ "input_cache_creation": 0.5e-6,+ "cache_write_tokens": 0.5e-6,+ "output": 1.5e-6,+ "output_reasoning_tokens": 1.5e-6,+ "output_reasoning": 1.5e-6,+ "reasoning_tokens": 1.5e-6+ }+ },+ {+ "id": "6f0606fb-b244-4a63-a0b4-08297b6c52a6",+ "name": "Flex Β· 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": 0.1e-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": 0.375e-6,+ "output_reasoning_tokens": 0.375e-6,+ "output_reasoning": 0.375e-6,+ "reasoning_tokens": 0.375e-6+ }+ },+ {+ "id": "32e5e542-0362-4471-ba8f-f2aa3aeb19dc",+ "name": "Fast mode",+ "isDefault": false,+ "priority": 3,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["fast", "priority"]+ }+ ],+ "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": 1e-6,+ "output_reasoning_tokens": 1e-6,+ "output_reasoning": 1e-6,+ "reasoning_tokens": 1e-6+ }+ },+ {+ "id": "405cdbb5-b5a1-4645-914d-f9f8bff5c3fd",+ "name": "Flex",+ "isDefault": false,+ "priority": 4,+ "conditions": [+ {+ "source": "model_parameters",+ "key": "service_tier",+ "operator": "in",+ "values": ["flex"]+ }+ ],+ "prices": {+ "input": 0.05e-6,+ "input_cached_tokens": 0.005e-6,+ "input_cache_read": 0.005e-6,+ "cache_read_input_tokens": 0.005e-6,+ "input_cache_creation": 0.0625e-6,+ "cache_write_tokens": 0.0625e-6,+ "output": 0.25e-6,+ "output_reasoning_tokens": 0.25e-6,+ "output_reasoning": 0.25e-6,+ "reasoning_tokens": 0.25e-6+ }+ },+ {+ "id": "8678a1a9-f622-4fce-93fe-77e258ca2b0a",+ "name": "Large Context (>272K)",+ "isDefault": false,+ "priority": 5,+ "conditions": [+ {+ "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": 0.75e-6,+ "output_reasoning_tokens": 0.75e-6,+ "output_reasoning": 0.75e-6,+ "reasoning_tokens": 0.75e-6+ }+ }+ ]+ },{"id": "6ed696e7-5f3a-4113-a233-019155fb3ff0","modelName": "gemini-3.5-flash",