From f6f0ab274d251c44fbd88844e1997e950ad19144 Mon Sep 17 00:00:00 2001
From: Patrick Wozniak <email@patwoz.de>
Date: Thu, 20 Aug 2026 00:13:56 +0200
Subject: [PATCH] fix(models): refresh DeepSeek V4 pricing

---
 src/pricing.ts                          | 17 +++++++++--------
 tests/fixtures/commandcode-pricing.json |  6 +++---
 tests/test-pricing.ts                   | 14 ++++++++++----
 3 files changed, 22 insertions(+), 15 deletions(-)

diff --git a/src/pricing.ts b/src/pricing.ts
index f13a69e..0dd9115 100644
--- a/src/pricing.ts
+++ b/src/pricing.ts
@@ -20,7 +20,7 @@ export interface TemporaryPricing {
 }
 
 export const PRICING_SOURCE_URL = "https://commandcode.ai/docs/resources/pricing-limits"
-export const PRICING_LAST_VERIFIED = "2026-08-18"
+export const PRICING_LAST_VERIFIED = "2026-08-20"
 
 export const ZERO_MODEL_COST: CommandCodeModelCost = {
   input: 0,
@@ -61,17 +61,18 @@ export const MODEL_COSTS: Readonly<Record<string, CommandCodeModelCost>> = {
   "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 V4 uses time-dependent rates. Display the documented off-peak
+  // rates, which apply for 17 hours per day; the Usage page remains authoritative.
   "deepseek/deepseek-v4-pro": {
-    input: 0.435,
-    output: 0.87,
-    cacheRead: 0.003625,
+    input: 0.66,
+    output: 1.98,
+    cacheRead: 0.022,
     cacheWrite: 0,
   },
   "deepseek/deepseek-v4-flash": {
-    input: 0.14,
-    output: 0.28,
-    cacheRead: 0.0028,
+    input: 0.22,
+    output: 0.66,
+    cacheRead: 0.007,
     cacheWrite: 0,
   },
   "Qwen/Qwen3.8-Max": { input: 2, output: 6, cacheRead: 0.25, cacheWrite: 2.5 },
diff --git a/tests/fixtures/commandcode-pricing.json b/tests/fixtures/commandcode-pricing.json
index 786cec2..6a508fd 100644
--- a/tests/fixtures/commandcode-pricing.json
+++ b/tests/fixtures/commandcode-pricing.json
@@ -1,5 +1,5 @@
 {
-  "verifiedAt": "2026-08-18",
+  "verifiedAt": "2026-08-20",
   "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": {
@@ -25,8 +25,8 @@
     "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],
+    "deepseek/deepseek-v4-pro": [0.66, 1.98, 0.022, 0],
+    "deepseek/deepseek-v4-flash": [0.22, 0.66, 0.007, 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],
diff --git a/tests/test-pricing.ts b/tests/test-pricing.ts
index c463fcf..6e17400 100644
--- a/tests/test-pricing.ts
+++ b/tests/test-pricing.ts
@@ -108,10 +108,16 @@ describe("MODEL_COSTS pricing overlay", () => {
   })
 
   it("matches corrected official rates", () => {
+    assertCost("deepseek/deepseek-v4-pro", {
+      input: 0.66,
+      output: 1.98,
+      cacheRead: 0.022,
+      cacheWrite: 0,
+    })
     assertCost("deepseek/deepseek-v4-flash", {
-      input: 0.14,
-      output: 0.28,
-      cacheRead: 0.0028,
+      input: 0.22,
+      output: 0.66,
+      cacheRead: 0.007,
       cacheWrite: 0,
     })
     assertCost("Qwen/Qwen3.7-Max", {
@@ -163,7 +169,7 @@ describe("MODEL_COSTS pricing overlay", () => {
 
   it("tracks pricing provenance", () => {
     assert.equal(PRICING_SOURCE_URL, "https://commandcode.ai/docs/resources/pricing-limits")
-    assert.equal(PRICING_LAST_VERIFIED, "2026-08-18")
+    assert.equal(PRICING_LAST_VERIFIED, "2026-08-20")
   })
 
   it("fails once temporary pricing needs review", () => {