fix(models): refresh model pricing

CHANGELOG.md

@@ -2,6 +2,7 @@
 
 ## Unreleased
 
+- Refresh display pricing for the current Command Code model catalog, remove expired Qwen promotional rates, add current free and discounted models, and require review when temporary prices expire.
 - Use the host-provided `pi-ai` and `pi-coding-agent` core packages instead of installing private runtime copies, including for local and out-of-store development checkouts.
 
 ## 0.4.4 - 2026-08-03

index.ts

@@ -20,57 +20,13 @@ import { COMMAND_CODE_CLI_VERSION, createStreamCommandCode, DEFAULT_API_BASE } f
 import { calculateCommandCodeCost } from "./src/cost.ts"
 import { DEFAULT_MODELS_URL, loadCommandCodeModels } from "./src/models.ts"
 import { getApiKey, login, refreshToken } from "./src/oauth.ts"
+import { MODEL_COSTS, ZERO_MODEL_COST } from "./src/pricing.ts"
 
 const API_BASE = process.env.COMMANDCODE_API_BASE ?? DEFAULT_API_BASE
 const MODELS_URL = process.env.COMMANDCODE_MODELS_URL ?? DEFAULT_MODELS_URL
 const MODELS_CACHE_PATH =
   process.env.COMMANDCODE_MODELS_CACHE ?? join(getAgentDir(), "commandcode-models.json")
 
-type CommandCodeModelCost = {
-  input: number
-  output: number
-  cacheRead: number
-  cacheWrite: number
-}
-
-const ZERO_MODEL_COST: CommandCodeModelCost = {
-  input: 0,
-  output: 0,
-  cacheRead: 0,
-  cacheWrite: 0,
-}
-
-// The Provider API supplies the current model list. Keep known display pricing
-// here until the Provider API exposes prices directly.
-const MODEL_COSTS: Record<string, CommandCodeModelCost> = {
-  "claude-opus-4-7": { input: 5, output: 25, cacheRead: 0.5, cacheWrite: 6.25 },
-  "claude-opus-4-6": { input: 5, output: 25, cacheRead: 0.5, cacheWrite: 6.25 },
-  "claude-sonnet-4-6": { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 3.75 },
-  "claude-haiku-4-5-20251001": { input: 1, output: 5, cacheRead: 0.1, cacheWrite: 1.25 },
-  "gpt-5.5": { input: 5, output: 30, cacheRead: 0.5, cacheWrite: 0 },
-  "gpt-5.4": { input: 2.5, output: 15, cacheRead: 0.25, cacheWrite: 0 },
-  "gpt-5.3-codex": { input: 2, output: 8, cacheRead: 0.5, cacheWrite: 0 },
-  "gpt-5.4-mini": { input: 0.75, output: 4.5, cacheRead: 0.075, cacheWrite: 0 },
-  "google/gemini-3.5-flash": { input: 1.5, output: 9, cacheRead: 0.15, cacheWrite: 0 },
-  "google/gemini-3.1-flash-lite": { input: 0.25, output: 1.5, cacheRead: 0.03, cacheWrite: 0 },
-  // 4× usage deal: 75% off (permanent, no expiry)
-  "deepseek/deepseek-v4-pro": { input: 0.435, output: 0.87, cacheRead: 0.003625, cacheWrite: 0 },
-  "deepseek/deepseek-v4-flash": { input: 0.14, output: 0.28, cacheRead: 0.028, cacheWrite: 0 },
-  "moonshotai/Kimi-K2.6": { input: 0.95, output: 4, cacheRead: 0.16, cacheWrite: 0 },
-  "moonshotai/Kimi-K2.5": { input: 0.6, output: 3, cacheRead: 0.1, cacheWrite: 0 },
-  "zai-org/GLM-5.1": { input: 1.4, output: 4.4, cacheRead: 0.26, cacheWrite: 0 },
-  "zai-org/GLM-5": { input: 1, output: 3.2, cacheRead: 0.2, cacheWrite: 0 },
-  "MiniMaxAI/MiniMax-M2.7": { input: 0.3, output: 1.2, cacheRead: 0.06, cacheWrite: 0 },
-  "MiniMaxAI/MiniMax-M2.5": { input: 0.27, output: 0.95, cacheRead: 0.03, cacheWrite: 0 },
-  "Qwen/Qwen3.6-Max-Preview": { input: 1.3, output: 7.8, cacheRead: 0.26, cacheWrite: 1.63 },
-  "Qwen/Qwen3.6-Plus": { input: 0.5, output: 3, cacheRead: 0.1, cacheWrite: 0 },
-  // 2× usage deal: 50% off through June 22, 2026
-  "Qwen/Qwen3.7-Max": { input: 1.25, output: 3.75, cacheRead: 0.25, cacheWrite: 1.56 },
-  "stepfun/Step-3.5-Flash": { input: 0.1, output: 0.3, cacheRead: 0.02, cacheWrite: 0 },
-  "xiaomi/mimo-v2.5-pro": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
-  "xiaomi/mimo-v2.5": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
-}
-
 const streamCommandCode = createStreamCommandCode({
   createStream: () => new AssistantMessageEventStream(),
   calculateCost: calculateCommandCodeCost,

src/cost.ts

@@ -10,10 +10,22 @@
 import type { ModelLike, Usage } from "./types.ts"
 
 export function calculateCommandCodeCost(model: ModelLike, usage: Usage): void {
-  usage.cost.input = (model.cost.input / 1_000_000) * usage.input
+  const inputTokens = usage.input + usage.cacheRead + usage.cacheWrite
-  usage.cost.output = (model.cost.output / 1_000_000) * usage.output
+  let rates = model.cost
-  usage.cost.cacheRead = (model.cost.cacheRead / 1_000_000) * usage.cacheRead
+  let matchedThreshold = -1
-  usage.cost.cacheWrite = (model.cost.cacheWrite * usage.cacheWrite) / 1_000_000
+  for (const tier of model.cost.tiers ?? []) {
+    if (inputTokens > tier.inputTokensAbove && tier.inputTokensAbove > matchedThreshold) {
+      rates = tier
+      matchedThreshold = tier.inputTokensAbove
+    }
+  }
+
+  const longWrite = usage.cacheWrite1h ?? 0
+  const shortWrite = usage.cacheWrite - longWrite
+  usage.cost.input = (rates.input / 1_000_000) * usage.input
+  usage.cost.output = (rates.output / 1_000_000) * usage.output
+  usage.cost.cacheRead = (rates.cacheRead / 1_000_000) * usage.cacheRead
+  usage.cost.cacheWrite = (rates.cacheWrite * shortWrite + rates.input * 2 * longWrite) / 1_000_000
   usage.cost.total =
     usage.cost.input + usage.cost.output + usage.cost.cacheRead + usage.cost.cacheWrite
 }

src/pricing.ts

@@ -0,0 +1,226 @@
+export interface CommandCodeModelCostRates {
+  input: number
+  output: number
+  cacheRead: number
+  cacheWrite: number
+}
+
+export interface CommandCodeModelCostTier extends CommandCodeModelCostRates {
+  inputTokensAbove: number
+}
+
+export interface CommandCodeModelCost extends CommandCodeModelCostRates {
+  tiers?: readonly CommandCodeModelCostTier[]
+}
+
+export interface TemporaryPricing {
+  models: readonly string[]
+  expiresOn: string
+  description: string
+}
+
+export const PRICING_SOURCE_URL = "https://commandcode.ai/docs/resources/pricing-limits"
+export const PRICING_LAST_VERIFIED = "2026-08-04"
+
+export const ZERO_MODEL_COST: CommandCodeModelCost = {
+  input: 0,
+  output: 0,
+  cacheRead: 0,
+  cacheWrite: 0,
+}
+
+/**
+ * Display prices in USD per million tokens.
+ *
+ * Context-dependent rates use pi's request-wide input pricing tiers. The
+ * highest threshold exceeded by input + cache reads + cache writes applies to
+ * the full request. The Command Code usage page remains authoritative for the
+ * amount billed for an individual request.
+ */
+export const MODEL_COSTS: Readonly<Record<string, CommandCodeModelCost>> = {
+  // Free models
+  "poolside/laguna-s-2.1-free": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
+  "inclusionai/ling-3.0-flash-free": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
+
+  // Open and open-weight models
+  "tencent/hy3-paid": { input: 0.14, output: 0.58, cacheRead: 0.035, cacheWrite: 0 },
+  "moonshotai/Kimi-K3": { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 0 },
+  "moonshotai/Kimi-K2.7-Code": { input: 0.95, output: 4, cacheRead: 0.19, cacheWrite: 0 },
+  "moonshotai/Kimi-K2.7-Code-Highspeed": {
+    input: 1.9,
+    output: 8,
+    cacheRead: 0.38,
+    cacheWrite: 0,
+  },
+  "moonshotai/Kimi-K2.6": { input: 0.95, output: 4, cacheRead: 0.16, cacheWrite: 0 },
+  "moonshotai/Kimi-K2.5": { input: 0.6, output: 3, cacheRead: 0.1, cacheWrite: 0 },
+  "zai-org/GLM-5.2": { input: 1.4, output: 4.4, cacheRead: 0.26, cacheWrite: 0 },
+  "zai-org/GLM-5.2-Fast": { input: 3, output: 10.25, cacheRead: 0.5, cacheWrite: 0 },
+  "zai-org/GLM-5.1": { input: 1.4, output: 4.4, cacheRead: 0.26, cacheWrite: 0 },
+  "zai-org/GLM-5": { input: 1, output: 3.2, cacheRead: 0.2, cacheWrite: 0 },
+  "MiniMaxAI/MiniMax-M3": { input: 0.3, output: 1.2, cacheRead: 0.06, cacheWrite: 0 },
+  "MiniMaxAI/MiniMax-M2.7": { input: 0.3, output: 1.2, cacheRead: 0.06, cacheWrite: 0 },
+  "MiniMaxAI/MiniMax-M2.5": { input: 0.3, output: 1.2, cacheRead: 0.03, cacheWrite: 0 },
+  // Permanent 75% discount.
+  "deepseek/deepseek-v4-pro": {
+    input: 0.435,
+    output: 0.87,
+    cacheRead: 0.003625,
+    cacheWrite: 0,
+  },
+  "deepseek/deepseek-v4-flash": {
+    input: 0.14,
+    output: 0.28,
+    cacheRead: 0.0028,
+    cacheWrite: 0,
+  },
+  "Qwen/Qwen3.8-Max": { input: 2, output: 6, cacheRead: 0.25, cacheWrite: 2.5 },
+  "Qwen/Qwen3.7-Max": { input: 2.5, output: 7.5, cacheRead: 0.5, cacheWrite: 3.13 },
+  "Qwen/Qwen3.7-Plus": {
+    input: 0.4,
+    output: 1.6,
+    cacheRead: 0.08,
+    cacheWrite: 0.5,
+    tiers: [
+      {
+        inputTokensAbove: 256_000,
+        input: 1.2,
+        output: 4.8,
+        cacheRead: 0.24,
+        cacheWrite: 1.5,
+      },
+    ],
+  },
+  "Qwen/Qwen3.7-Flash": {
+    input: 0.03,
+    output: 0.13,
+    cacheRead: 0.006,
+    cacheWrite: 0.038,
+    tiers: [
+      {
+        inputTokensAbove: 32_000,
+        input: 0.1,
+        output: 0.4,
+        cacheRead: 0.02,
+        cacheWrite: 0.125,
+      },
+      {
+        inputTokensAbove: 256_000,
+        input: 0.2,
+        output: 0.8,
+        cacheRead: 0.04,
+        cacheWrite: 0.25,
+      },
+    ],
+  },
+  "Qwen/Qwen3.6-Max-Preview": {
+    input: 1.3,
+    output: 7.8,
+    cacheRead: 0.26,
+    cacheWrite: 1.63,
+  },
+  "Qwen/Qwen3.6-Plus": { input: 0.5, output: 3, cacheRead: 0.1, cacheWrite: 0 },
+  "stepfun/Step-3.7-Flash": { input: 0.2, output: 1.15, cacheRead: 0.04, cacheWrite: 0 },
+  "stepfun/Step-3.5-Flash": { input: 0.1, output: 0.3, cacheRead: 0.02, cacheWrite: 0 },
+  // Permanent discounted rates.
+  "xiaomi/mimo-v2.5-pro": { input: 0.435, output: 0.87, cacheRead: 0.0036, cacheWrite: 0 },
+  "xiaomi/mimo-v2.5": { input: 0.14, output: 0.28, cacheRead: 0.0028, cacheWrite: 0 },
+  "nvidia/nemotron-3-ultra-550b-a55b": {
+    input: 0.6,
+    output: 2.4,
+    cacheRead: 0.12,
+    cacheWrite: 0,
+  },
+  "sakana/fugu-ultra": { input: 5, output: 30, cacheRead: 0.5, cacheWrite: 0 },
+  "thinkingmachines/inkling": { input: 1, output: 4.05, cacheRead: 0.17, cacheWrite: 0 },
+  "thinkingmachines/inkling-small": {
+    input: 0.5,
+    output: 1.2,
+    cacheRead: 0.1,
+    cacheWrite: 0,
+  },
+  "meta/muse-spark-1.1": { input: 1.25, output: 4.25, cacheRead: 0.15, cacheWrite: 0 },
+
+  // Anthropic
+  // Introductory pricing through 2026-08-31.
+  "claude-sonnet-5": { input: 2, output: 10, cacheRead: 0.2, cacheWrite: 2.5 },
+  "claude-sonnet-4-6": { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 3.75 },
+  "claude-fable-5": { input: 10, output: 50, cacheRead: 1, cacheWrite: 12.5 },
+  "claude-opus-5": { input: 5, output: 25, cacheRead: 0.5, cacheWrite: 6.25 },
+  "claude-opus-4-8": { input: 5, output: 25, cacheRead: 0.5, cacheWrite: 6.25 },
+  "claude-opus-4-7": { input: 5, output: 25, cacheRead: 0.5, cacheWrite: 6.25 },
+  "claude-haiku-4-5-20251001": {
+    input: 1,
+    output: 5,
+    cacheRead: 0.1,
+    cacheWrite: 1.25,
+  },
+
+  // OpenAI
+  "gpt-5.6-sol": { input: 5, output: 30, cacheRead: 0.5, cacheWrite: 6.25 },
+  // Discounted rates through 2026-08-14.
+  "gpt-5.6-terra": {
+    input: 1,
+    output: 6,
+    cacheRead: 0.1,
+    cacheWrite: 1.25,
+    tiers: [
+      {
+        inputTokensAbove: 272_000,
+        input: 2,
+        output: 9,
+        cacheRead: 0.2,
+        cacheWrite: 2.5,
+      },
+    ],
+  },
+  "gpt-5.6-luna": {
+    input: 0.1,
+    output: 0.6,
+    cacheRead: 0.01,
+    cacheWrite: 0.125,
+    tiers: [
+      {
+        inputTokensAbove: 272_000,
+        input: 0.2,
+        output: 0.9,
+        cacheRead: 0.02,
+        cacheWrite: 0.25,
+      },
+    ],
+  },
+  "gpt-5.5": { input: 5, output: 30, cacheRead: 0.5, cacheWrite: 0 },
+  "gpt-5.4": { input: 2.5, output: 15, cacheRead: 0.25, cacheWrite: 0 },
+  "gpt-5.3-codex": { input: 2, output: 8, cacheRead: 0.5, cacheWrite: 0 },
+  "gpt-5.4-mini": { input: 0.75, output: 4.5, cacheRead: 0.075, cacheWrite: 0 },
+
+  // Google and xAI
+  "google/gemini-3.6-flash": { input: 1.5, output: 7.5, cacheRead: 0.15, cacheWrite: 0 },
+  "google/gemini-3.5-flash": { input: 1.5, output: 9, cacheRead: 0.15, cacheWrite: 0 },
+  "google/gemini-3.5-flash-lite": {
+    input: 0.3,
+    output: 2.5,
+    cacheRead: 0.03,
+    cacheWrite: 0,
+  },
+  "google/gemini-3.1-flash-lite": {
+    input: 0.25,
+    output: 1.5,
+    cacheRead: 0.03,
+    cacheWrite: 0,
+  },
+  "xai/grok-4.5": { input: 2, output: 6, cacheRead: 0.5, cacheWrite: 0 },
+}
+
+export const TEMPORARY_PRICING: readonly TemporaryPricing[] = [
+  {
+    models: ["gpt-5.6-terra", "gpt-5.6-luna"],
+    expiresOn: "2026-08-14",
+    description: "50% promotional rates",
+  },
+  {
+    models: ["claude-sonnet-5"],
+    expiresOn: "2026-08-31",
+    description: "introductory pricing",
+  },
+]

src/types.ts

@@ -15,6 +15,7 @@ export interface Usage {
   output: number
   cacheRead: number
   cacheWrite: number
+  cacheWrite1h?: number
   totalTokens: number
   cost: UsageCost
 }
@@ -50,13 +51,21 @@ export interface AssistantMessageLike {
   timestamp: number
 }
 
-export interface ModelCost {
+export interface ModelCostRates {
   input: number
   output: number
   cacheRead: number
   cacheWrite: number
 }
 
+export interface ModelCostTier extends ModelCostRates {
+  inputTokensAbove: number
+}
+
+export interface ModelCost extends ModelCostRates {
+  tiers?: readonly ModelCostTier[]
+}
+
 export interface ModelLike {
   id: string
   api: unknown

tests/fixtures/commandcode-model-ids.json

@@ -0,0 +1,57 @@
+{
+  "fetchedAt": "2026-08-04T10:12:57.953Z",
+  "source": "https://api.commandcode.ai/provider/v1/models",
+  "modelIds": [
+    "claude-sonnet-5",
+    "claude-sonnet-4-6",
+    "claude-fable-5",
+    "claude-opus-5",
+    "claude-opus-4-8",
+    "claude-opus-4-7",
+    "claude-haiku-4-5-20251001",
+    "gpt-5.6-sol",
+    "gpt-5.6-terra",
+    "gpt-5.6-luna",
+    "gpt-5.5",
+    "gpt-5.4",
+    "gpt-5.3-codex",
+    "gpt-5.4-mini",
+    "deepseek/deepseek-v4-pro",
+    "deepseek/deepseek-v4-flash",
+    "moonshotai/Kimi-K3",
+    "moonshotai/Kimi-K2.7-Code",
+    "moonshotai/Kimi-K2.7-Code-Highspeed",
+    "moonshotai/Kimi-K2.6",
+    "moonshotai/Kimi-K2.5",
+    "zai-org/GLM-5.2",
+    "zai-org/GLM-5.2-Fast",
+    "zai-org/GLM-5.1",
+    "zai-org/GLM-5",
+    "MiniMaxAI/MiniMax-M3",
+    "MiniMaxAI/MiniMax-M2.7",
+    "MiniMaxAI/MiniMax-M2.5",
+    "xiaomi/mimo-v2.5-pro",
+    "xiaomi/mimo-v2.5",
+    "Qwen/Qwen3.8-Max",
+    "Qwen/Qwen3.7-Max",
+    "Qwen/Qwen3.7-Plus",
+    "Qwen/Qwen3.7-Flash",
+    "Qwen/Qwen3.6-Max-Preview",
+    "Qwen/Qwen3.6-Plus",
+    "stepfun/Step-3.7-Flash",
+    "stepfun/Step-3.5-Flash",
+    "tencent/hy3-paid",
+    "google/gemini-3.6-flash",
+    "google/gemini-3.5-flash",
+    "google/gemini-3.5-flash-lite",
+    "google/gemini-3.1-flash-lite",
+    "sakana/fugu-ultra",
+    "nvidia/nemotron-3-ultra-550b-a55b",
+    "thinkingmachines/inkling",
+    "thinkingmachines/inkling-small",
+    "poolside/laguna-s-2.1-free",
+    "inclusionai/ling-3.0-flash-free",
+    "meta/muse-spark-1.1",
+    "xai/grok-4.5"
+  ]
+}

tests/fixtures/commandcode-pricing.json

@@ -0,0 +1,67 @@
+{
+  "verifiedAt": "2026-08-04",
+  "source": "https://commandcode.ai/docs/resources/pricing-limits",
+  "tierPolicy": "Use request-wide input tiers; the highest threshold exceeded by input plus cache tokens applies to the full request.",
+  "tiers": {
+    "Qwen/Qwen3.7-Plus": [[256000, 1.2, 4.8, 0.24, 1.5]],
+    "Qwen/Qwen3.7-Flash": [
+      [32000, 0.1, 0.4, 0.02, 0.125],
+      [256000, 0.2, 0.8, 0.04, 0.25]
+    ],
+    "gpt-5.6-terra": [[272000, 2, 9, 0.2, 2.5]],
+    "gpt-5.6-luna": [[272000, 0.2, 0.9, 0.02, 0.25]]
+  },
+  "costs": {
+    "poolside/laguna-s-2.1-free": [0, 0, 0, 0],
+    "inclusionai/ling-3.0-flash-free": [0, 0, 0, 0],
+    "tencent/hy3-paid": [0.14, 0.58, 0.035, 0],
+    "moonshotai/Kimi-K3": [3, 15, 0.3, 0],
+    "moonshotai/Kimi-K2.7-Code": [0.95, 4, 0.19, 0],
+    "moonshotai/Kimi-K2.7-Code-Highspeed": [1.9, 8, 0.38, 0],
+    "moonshotai/Kimi-K2.6": [0.95, 4, 0.16, 0],
+    "moonshotai/Kimi-K2.5": [0.6, 3, 0.1, 0],
+    "zai-org/GLM-5.2": [1.4, 4.4, 0.26, 0],
+    "zai-org/GLM-5.2-Fast": [3, 10.25, 0.5, 0],
+    "zai-org/GLM-5.1": [1.4, 4.4, 0.26, 0],
+    "zai-org/GLM-5": [1, 3.2, 0.2, 0],
+    "MiniMaxAI/MiniMax-M3": [0.3, 1.2, 0.06, 0],
+    "MiniMaxAI/MiniMax-M2.7": [0.3, 1.2, 0.06, 0],
+    "MiniMaxAI/MiniMax-M2.5": [0.3, 1.2, 0.03, 0],
+    "deepseek/deepseek-v4-pro": [0.435, 0.87, 0.003625, 0],
+    "deepseek/deepseek-v4-flash": [0.14, 0.28, 0.0028, 0],
+    "Qwen/Qwen3.8-Max": [2, 6, 0.25, 2.5],
+    "Qwen/Qwen3.7-Max": [2.5, 7.5, 0.5, 3.13],
+    "Qwen/Qwen3.7-Plus": [0.4, 1.6, 0.08, 0.5],
+    "Qwen/Qwen3.7-Flash": [0.03, 0.13, 0.006, 0.038],
+    "Qwen/Qwen3.6-Max-Preview": [1.3, 7.8, 0.26, 1.63],
+    "Qwen/Qwen3.6-Plus": [0.5, 3, 0.1, 0],
+    "stepfun/Step-3.7-Flash": [0.2, 1.15, 0.04, 0],
+    "stepfun/Step-3.5-Flash": [0.1, 0.3, 0.02, 0],
+    "xiaomi/mimo-v2.5-pro": [0.435, 0.87, 0.0036, 0],
+    "xiaomi/mimo-v2.5": [0.14, 0.28, 0.0028, 0],
+    "nvidia/nemotron-3-ultra-550b-a55b": [0.6, 2.4, 0.12, 0],
+    "sakana/fugu-ultra": [5, 30, 0.5, 0],
+    "thinkingmachines/inkling": [1, 4.05, 0.17, 0],
+    "thinkingmachines/inkling-small": [0.5, 1.2, 0.1, 0],
+    "meta/muse-spark-1.1": [1.25, 4.25, 0.15, 0],
+    "claude-sonnet-5": [2, 10, 0.2, 2.5],
+    "claude-sonnet-4-6": [3, 15, 0.3, 3.75],
+    "claude-fable-5": [10, 50, 1, 12.5],
+    "claude-opus-5": [5, 25, 0.5, 6.25],
+    "claude-opus-4-8": [5, 25, 0.5, 6.25],
+    "claude-opus-4-7": [5, 25, 0.5, 6.25],
+    "claude-haiku-4-5-20251001": [1, 5, 0.1, 1.25],
+    "gpt-5.6-sol": [5, 30, 0.5, 6.25],
+    "gpt-5.6-terra": [1, 6, 0.1, 1.25],
+    "gpt-5.6-luna": [0.1, 0.6, 0.01, 0.125],
+    "gpt-5.5": [5, 30, 0.5, 0],
+    "gpt-5.4": [2.5, 15, 0.25, 0],
+    "gpt-5.3-codex": [2, 8, 0.5, 0],
+    "gpt-5.4-mini": [0.75, 4.5, 0.075, 0],
+    "google/gemini-3.6-flash": [1.5, 7.5, 0.15, 0],
+    "google/gemini-3.5-flash": [1.5, 9, 0.15, 0],
+    "google/gemini-3.5-flash-lite": [0.3, 2.5, 0.03, 0],
+    "google/gemini-3.1-flash-lite": [0.25, 1.5, 0.03, 0],
+    "xai/grok-4.5": [2, 6, 0.5, 0]
+  }
+}

tests/helpers.ts

@@ -79,7 +79,7 @@ export function makeModel(overrides: Partial<ModelLike> = {}): ModelLike {
     api: "commandcode-custom",
     provider: "commandcode",
     maxTokens: 384_000,
-    cost: { input: 0.14, output: 0.28, cacheRead: 0.028, cacheWrite: 0 },
+    cost: { input: 0.14, output: 0.28, cacheRead: 0.0028, cacheWrite: 0 },
     ...overrides,
   }
 }

tests/test-cost.ts

@@ -13,13 +13,17 @@ import { describe, it } from "node:test"
 import { calculateCommandCodeCost } from "../src/cost.ts"
 import type { Usage } from "../src/types.ts"
 
-interface CostTable {
+interface CostRates {
   input: number
   output: number
   cacheRead: number
   cacheWrite: number
 }
 
+interface CostTable extends CostRates {
+  tiers?: Array<CostRates & { inputTokensAbove: number }>
+}
+
 const COST_FIXTURES: Record<string, CostTable> = {
   "zero-cost-model": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
   "claude-sonnet-4-6": { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 3.75 },
@@ -29,7 +33,17 @@ const COST_FIXTURES: Record<string, CostTable> = {
     cacheRead: 0.003625,
     cacheWrite: 0,
   },
-  "Qwen/Qwen3.7-Max": { input: 1.25, output: 3.75, cacheRead: 0.25, cacheWrite: 1.56 },
+  "Qwen/Qwen3.7-Max": { input: 2.5, output: 7.5, cacheRead: 0.5, cacheWrite: 3.13 },
+  "Qwen/Qwen3.7-Flash": {
+    input: 0.03,
+    output: 0.13,
+    cacheRead: 0.006,
+    cacheWrite: 0.038,
+    tiers: [
+      { inputTokensAbove: 32_000, input: 0.1, output: 0.4, cacheRead: 0.02, cacheWrite: 0.125 },
+      { inputTokensAbove: 256_000, input: 0.2, output: 0.8, cacheRead: 0.04, cacheWrite: 0.25 },
+    ],
+  },
   "gpt-5.5": { input: 5, output: 30, cacheRead: 0.5, cacheWrite: 0 },
 }
 
@@ -51,6 +65,14 @@ function commandCodeModel(id: string, cost: CostTable) {
   }
 }
 
+function assertClose(actual: number, expected: number) {
+  assert.ok(
+    Math.abs(actual - expected) <=
+      Number.EPSILON * Math.max(1, Math.abs(actual), Math.abs(expected)),
+    `expected ${actual} to be close to ${expected}`,
+  )
+}
+
 function freshUsage(tokens: (typeof USAGE_CASES)[number]): Usage {
   return {
     ...tokens,
@@ -60,10 +82,20 @@ function freshUsage(tokens: (typeof USAGE_CASES)[number]): Usage {
 }
 
 function expectedCost(cost: CostTable, tokens: (typeof USAGE_CASES)[number]): Usage["cost"] {
-  const input = (cost.input / 1_000_000) * tokens.input
+  const inputTokens = tokens.input + tokens.cacheRead + tokens.cacheWrite
-  const output = (cost.output / 1_000_000) * tokens.output
+  let rates: CostRates = cost
-  const cacheRead = (cost.cacheRead / 1_000_000) * tokens.cacheRead
+  let matchedThreshold = -1
-  const cacheWrite = (cost.cacheWrite * tokens.cacheWrite) / 1_000_000
+  for (const tier of cost.tiers ?? []) {
+    if (inputTokens > tier.inputTokensAbove && tier.inputTokensAbove > matchedThreshold) {
+      rates = tier
+      matchedThreshold = tier.inputTokensAbove
+    }
+  }
+
+  const input = (rates.input / 1_000_000) * tokens.input
+  const output = (rates.output / 1_000_000) * tokens.output
+  const cacheRead = (rates.cacheRead / 1_000_000) * tokens.cacheRead
+  const cacheWrite = (rates.cacheWrite * tokens.cacheWrite) / 1_000_000
   return {
     input,
     output,
@@ -91,6 +123,51 @@ describe("calculateCommandCodeCost()", () => {
     }
   })
 
+  it("applies the highest request-wide input tier above its threshold", () => {
+    const model = commandCodeModel("Qwen/Qwen3.7-Flash", COST_FIXTURES["Qwen/Qwen3.7-Flash"])
+
+    const atThreshold = freshUsage({
+      input: 32_000,
+      output: 1_000,
+      cacheRead: 0,
+      cacheWrite: 0,
+    })
+    calculateCommandCodeCost(model, atThreshold)
+    assertClose(atThreshold.cost.input, (0.03 * 32_000) / 1_000_000)
+
+    const aboveFirstTier = freshUsage({
+      input: 30_000,
+      output: 1_000,
+      cacheRead: 2_001,
+      cacheWrite: 0,
+    })
+    calculateCommandCodeCost(model, aboveFirstTier)
+    assertClose(aboveFirstTier.cost.input, (0.1 * 30_000) / 1_000_000)
+    assertClose(aboveFirstTier.cost.cacheRead, (0.02 * 2_001) / 1_000_000)
+
+    const aboveHighestTier = freshUsage({
+      input: 100_000,
+      output: 1_000,
+      cacheRead: 156_001,
+      cacheWrite: 0,
+    })
+    calculateCommandCodeCost(model, aboveHighestTier)
+    assertClose(aboveHighestTier.cost.input, (0.2 * 100_000) / 1_000_000)
+    assertClose(aboveHighestTier.cost.output, (0.8 * 1_000) / 1_000_000)
+  })
+
+  it("prices one-hour cache writes at twice the active input rate", () => {
+    const model = commandCodeModel("claude-sonnet-4-6", COST_FIXTURES["claude-sonnet-4-6"])
+    const usage = freshUsage({ input: 0, output: 0, cacheRead: 0, cacheWrite: 1_000 })
+    usage.cacheWrite1h = 400
+
+    calculateCommandCodeCost(model, usage)
+
+    const expectedShortWrite = (3.75 * 600) / 1_000_000
+    const expectedLongWrite = (3 * 2 * 400) / 1_000_000
+    assertClose(usage.cost.cacheWrite, expectedShortWrite + expectedLongWrite)
+  })
+
   it("writes the total as the sum of all cost components", () => {
     const model = commandCodeModel("claude-sonnet-4-6", COST_FIXTURES["claude-sonnet-4-6"])
     const usage = freshUsage({ input: 1_000, output: 500, cacheRead: 10_000, cacheWrite: 2_000 })

tests/test-pricing.ts

@@ -1,90 +1,177 @@
 import assert from "node:assert/strict"
+import { readFile } from "node:fs/promises"
 import { describe, it } from "node:test"
 
-// MODEL_COSTS is a module-level const in index.ts. We verify the pricing
+import {
-// overlay by importing the map through a dedicated re-export so tests don't
+  MODEL_COSTS,
-// need to spin up the full extension.
+  PRICING_LAST_VERIFIED,
-//
+  PRICING_SOURCE_URL,
-// To keep the test self-contained without importing the full extension (which
+  TEMPORARY_PRICING,
-// requires ExtensionAPI), we read the source and extract the constant at
+} from "../src/pricing.ts"
-// runtime. A cleaner approach would be a dedicated src/pricing.ts module,
-// but for now we verify the known cost entries directly.
 
-import { readFileSync } from "node:fs"
+interface ModelCatalogSnapshot {
-import { resolve, dirname } from "node:path"
+  fetchedAt: string
-import { fileURLToPath } from "node:url"
+  source: string
+  modelIds: string[]
+}
 
-const __dirname = dirname(fileURLToPath(import.meta.url))
+interface PricingSnapshot {
-const indexSource = readFileSync(resolve(__dirname, "..", "index.ts"), "utf-8")
+  verifiedAt: string
+  source: string
+  tierPolicy: string
+  tiers: Record<string, [number, number, number, number, number][]>
+  costs: Record<string, [number, number, number, number]>
+}
 
-// Extract MODEL_COSTS object from index.ts source using a simple parse.
+const fixtureUrl = new URL("./fixtures/commandcode-model-ids.json", import.meta.url)
-// The map is written as a Record<string, {input:number,output:number,...}>
+const fixture = JSON.parse(await readFile(fixtureUrl, "utf-8")) as ModelCatalogSnapshot
-// so we eval it in a sandboxed context.
+const pricingFixtureUrl = new URL("./fixtures/commandcode-pricing.json", import.meta.url)
-const match = indexSource.match(
+const pricingFixture = JSON.parse(await readFile(pricingFixtureUrl, "utf-8")) as PricingSnapshot
-  /const MODEL_COSTS:\s*Record<string,\s*CommandCodeModelCost>\s*=\s*\{([\s\S]*?)\n\}/,
+const freeModels = new Set(["poolside/laguna-s-2.1-free", "inclusionai/ling-3.0-flash-free"])
+
+function assertCost(
+  modelId: string,
+  expected: { input: number; output: number; cacheRead: number; cacheWrite: number },
+) {
+  const cost = MODEL_COSTS[modelId]
+  assert.ok(cost, `${modelId} should have pricing`)
+  assert.deepEqual(
+    {
+      input: cost.input,
+      output: cost.output,
+      cacheRead: cost.cacheRead,
+      cacheWrite: cost.cacheWrite,
+    },
+    expected,
+    `${modelId} base pricing should match the source`,
   )
-assert.ok(match, "MODEL_COSTS constant should exist in index.ts")
-
-// Parse the cost entries from the extracted block.
-const costBlock = match[1]
-const entries: Record<string, { input: number; output: number }> = {}
-for (const line of costBlock.split("\n")) {
-  const trimmed = line.trim()
-  if (!trimmed || trimmed.startsWith("//")) continue
-  const entryMatch = trimmed.match(/^"([^"]+)":\s*\{\s*input:\s*([\d.]+),\s*output:\s*([\d.]+)/)
-  if (entryMatch) {
-    entries[entryMatch[1]] = {
-      input: Number(entryMatch[2]),
-      output: Number(entryMatch[3]),
-    }
-  }
 }
 
 describe("MODEL_COSTS pricing overlay", () => {
-  it("covers known Command Code models with non-zero pricing", () => {
+  it("covers the current Command Code model catalog snapshot", () => {
-    const knownModels = [
+    assert.equal(fixture.source, "https://api.commandcode.ai/provider/v1/models")
-      "deepseek/deepseek-v4-flash",
+    assert.match(fixture.fetchedAt, /^2026-08-04T/)
-      "deepseek/deepseek-v4-pro",
-      "claude-sonnet-4-6",
-      "claude-opus-4-7",
-      "Qwen/Qwen3.7-Max",
-      "gpt-5.5",
-      "stepfun/Step-3.5-Flash",
-    ]
 
-    for (const id of knownModels) {
+    const catalogIds = [...fixture.modelIds].sort()
-      const cost = entries[id]
+    const pricedIds = Object.keys(MODEL_COSTS).sort()
-      assert.ok(cost, `MODEL_COSTS should include "${id}"`)
+    assert.deepEqual(pricedIds, catalogIds)
-      assert.ok(cost.input > 0, `"${id}" input cost should be > 0`)
+  })
-      assert.ok(cost.output > 0, `"${id}" output cost should be > 0`)
+
+  it("matches the verified official pricing snapshot", () => {
+    assert.equal(pricingFixture.verifiedAt, PRICING_LAST_VERIFIED)
+    assert.equal(pricingFixture.source, PRICING_SOURCE_URL)
+    assert.match(pricingFixture.tierPolicy, /request-wide input tiers/)
+
+    const expected = Object.fromEntries(
+      Object.entries(pricingFixture.costs).map(
+        ([modelId, [input, output, cacheRead, cacheWrite]]) => [
+          modelId,
+          {
+            input,
+            output,
+            cacheRead,
+            cacheWrite,
+            ...(pricingFixture.tiers[modelId]
+              ? {
+                  tiers: pricingFixture.tiers[modelId].map(
+                    ([inputTokensAbove, tierInput, tierOutput, tierCacheRead, tierCacheWrite]) => ({
+                      inputTokensAbove,
+                      input: tierInput,
+                      output: tierOutput,
+                      cacheRead: tierCacheRead,
+                      cacheWrite: tierCacheWrite,
+                    }),
+                  ),
                 }
-  })
+              : {}),
-
+          },
-  it("includes promotional pricing notes in comments", () => {
+        ],
-    // The DeepSeek V4 Pro 4× deal and Qwen 3.7 Max 2× deal should be
-    // documented in the source comments.
-    assert.ok(
-      costBlock.includes("4× usage deal") || costBlock.includes("75% off"),
-      "DeepSeek V4 Pro promotional pricing should be documented",
-    )
-    assert.ok(
-      costBlock.includes("2× usage deal") || costBlock.includes("50% off"),
-      "Qwen 3.7 Max promotional pricing should be documented",
-    )
-  })
-
-  it("has cache pricing for models that support it", () => {
-    // Claude models should have non-zero cacheRead and cacheWrite costs.
-    const claudeModels = ["claude-sonnet-4-6", "claude-opus-4-7"]
-    for (const id of claudeModels) {
-      const fullEntryMatch = costBlock.match(
-        new RegExp(
-          `"${id.replace(/\//g, "\\\\")}":\\s*\\{[^}]+cacheRead:\\s*([\\d.]+)[^}]+cacheWrite:\\s*([\\d.]+)`,
       ),
     )
-      assert.ok(fullEntryMatch, `"${id}" should have cacheRead and cacheWrite fields`)
+    assert.deepEqual(MODEL_COSTS, expected)
-      assert.ok(Number(fullEntryMatch[1]) > 0, `"${id}" cacheRead should be > 0`)
+  })
-      assert.ok(Number(fullEntryMatch[2]) > 0, `"${id}" cacheWrite should be > 0`)
+
+  it("uses non-zero prices except for models documented as free", () => {
+    for (const [modelId, cost] of Object.entries(MODEL_COSTS)) {
+      assert.ok(cost.input >= 0, `${modelId} input cost should be non-negative`)
+      assert.ok(cost.output >= 0, `${modelId} output cost should be non-negative`)
+      assert.ok(cost.cacheRead >= 0, `${modelId} cache-read cost should be non-negative`)
+      assert.ok(cost.cacheWrite >= 0, `${modelId} cache-write cost should be non-negative`)
+
+      const allZero = Object.values(cost).every((value) => value === 0)
+      assert.equal(
+        allZero,
+        freeModels.has(modelId),
+        `${modelId} free-model status should be explicit`,
+      )
+    }
+  })
+
+  it("matches corrected official rates", () => {
+    assertCost("deepseek/deepseek-v4-flash", {
+      input: 0.14,
+      output: 0.28,
+      cacheRead: 0.0028,
+      cacheWrite: 0,
+    })
+    assertCost("Qwen/Qwen3.7-Max", {
+      input: 2.5,
+      output: 7.5,
+      cacheRead: 0.5,
+      cacheWrite: 3.13,
+    })
+    assertCost("xiaomi/mimo-v2.5-pro", {
+      input: 0.435,
+      output: 0.87,
+      cacheRead: 0.0036,
+      cacheWrite: 0,
+    })
+    assertCost("MiniMaxAI/MiniMax-M2.5", {
+      input: 0.3,
+      output: 1.2,
+      cacheRead: 0.03,
+      cacheWrite: 0,
+    })
+  })
+
+  it("uses the documented base rates for context-dependent models", () => {
+    assertCost("Qwen/Qwen3.7-Plus", {
+      input: 0.4,
+      output: 1.6,
+      cacheRead: 0.08,
+      cacheWrite: 0.5,
+    })
+    assertCost("Qwen/Qwen3.7-Flash", {
+      input: 0.03,
+      output: 0.13,
+      cacheRead: 0.006,
+      cacheWrite: 0.038,
+    })
+    assertCost("gpt-5.6-terra", {
+      input: 1,
+      output: 6,
+      cacheRead: 0.1,
+      cacheWrite: 1.25,
+    })
+  })
+
+  it("tracks pricing provenance", () => {
+    assert.equal(PRICING_SOURCE_URL, "https://commandcode.ai/docs/resources/pricing-limits")
+    assert.equal(PRICING_LAST_VERIFIED, "2026-08-04")
+  })
+
+  it("fails once temporary pricing needs review", () => {
+    const today = new Date().toISOString().slice(0, 10)
+    for (const pricing of TEMPORARY_PRICING) {
+      assert.match(pricing.expiresOn, /^\d{4}-\d{2}-\d{2}$/)
+      assert.ok(pricing.models.length > 0)
+      assert.ok(
+        pricing.expiresOn >= today,
+        `${pricing.description} for ${pricing.models.join(", ")} expired on ${pricing.expiresOn}; refresh MODEL_COSTS`,
+      )
+      for (const modelId of pricing.models) {
+        assert.ok(MODEL_COSTS[modelId], `${modelId} should have a temporary price entry`)
+      }
     }
   })
 })