feat(sse): restore GPT-6 Astra effort aliases and Codex 0.153.4 pin (#13026)

Three related things that genuinely belong together: the registry and suffix splitter not knowing Astra while the live catalog serves it, the Codex client pin Astra requires, and dropping `next/font/google` so a production image build does not reach fonts.googleapis.com. --- Validated in one consolidated worktree cut from `release/v3.8.51`, boarded with the other 19 PRs of this batch. Two in-batch conflicts, both additive and resolved by keeping each side: the `ENVIRONMENT.md` table (#13035 + #13011) and the `chatHelpers.ts` import block (#12975 on the tip + #13017). - `typecheck:core` clean; `check:dashboard-typecheck` OK (206 pre-existing, within baseline); `check:changelog-integrity` OK; `check:docs-counts` migrations ✓ - complexity 2816 / baseline 3218 and cognitive-complexity 1271 / baseline 1437 — both under baseline - 531 of 532 focused assertions green across the batch's 46 test files - `check-file-size` rebaselined for the batch's real growth (annotation `_rebaseline_2026_09_11_mergebatch_v3851_houminxi`, landed on #13038), attributed per PR The single red is **not this batch**: `tests/unit/combo/quota-weighted-strategy.test.ts` → "A/B isolation: 7 hard-empty + 2 at 0.5% + 1 at 40%, floor=1" asserts an order between two connections of identical weight and flakes on the pure tip too — 2 failures in 4 runs at `origin/release/v3.8.51` with nothing from this batch applied. ⚠️ base-red inherited: #12732 — `Docs Gates`, `Merge integrity`, `No new ESLint warnings`, `Unit Tests fast-path` and `Fast Quality Gates` reproduce on the pure tip (provider count 356 vs the 358 the modules define, SKILL.md drift, and `open-sse/utils/stream.ts` at 3115 > frozen 3098, untouched here). Thanks @HouMinXi — the live evidence on these (X500 logs, `storage.sqlite` state, real `/v1/models` probes, the 36-minute outage write-up) is what let a 20-PR batch be reviewed as a unit.
2026-09-13 18:32:12 +03:00 · 2026-09-11 18:28:25 -04:00
parent f4fd9190b3
commit d92bc8ef7a
25 changed files with 333 additions and 51 deletions
--- a/.env.example
+++ b/.env.example
@@ -1311,7 +1311,8 @@ CLAUDE_USER_AGENT="claude-cli/2.1.258 (external, cli)"
 # stream with a misleading 400 out-of-extra-usage placeholder. Set to true to
 # forward the original names verbatim (debugging only).
 # CLAUDE_DISABLE_TOOL_NAME_CLOAK=false
-CODEX_USER_AGENT="codex-cli/0.144.1 (Windows 10.0.26200; x64)"
+# Optional override; leave unset to follow the shared Codex client version.
+# CODEX_USER_AGENT="codex-cli/0.153.4 (Windows 10.0.26200; x64)"
 GITHUB_USER_AGENT="GitHubCopilotChat/0.54.0"
 ANTIGRAVITY_USER_AGENT="antigravity/2.0.1 linux/arm64 google-api-nodejs-client/10.3.0"
 KIRO_USER_AGENT="AWS-SDK-JS/3.0.0 kiro-ide/1.0.0"
@@ -1331,7 +1332,7 @@ CURSOR_USER_AGENT="Cursor/3.4"

 # Override Codex client version sent in headers independently of the
 # CODEX_USER_AGENT string. Used by: open-sse/config/codexClient.ts.
-# CODEX_CLIENT_VERSION=0.144.1
+# CODEX_CLIENT_VERSION=0.153.4
 #
 # Override the advertised Claude Code client version independently of
 # CLAUDE_USER_AGENT. Anthropic gates some models (Fable 5.1) on this
--- a/2
+++ b/2
@@ -340,7 +340,7 @@ RUN --mount=type=cache,id=s/92ca8a61-c1ba-421f-a389-d48ac7258c2d-apt-cache,targe
 #      build, not the floating `@latest`.
 RUN --mount=type=cache,id=s/92ca8a61-c1ba-421f-a389-d48ac7258c2d-npm-cache,target=/root/.npm \
  npm install -g --no-audit --no-fund \
-    @openai/codex@0.153.2 \
+    @openai/codex@0.153.4 \
    @anthropic-ai/claude-code@2.1.260 \
    droid@0.212.0 \
    openclaw@2026.9.1
--- a/changelog.d/features/codex-gpt-6-astra.md
+++ b/changelog.d/features/codex-gpt-6-astra.md
@@ -0,0 +1 @@
+- **feat(sse):** Codex and OpenAI catalogs list GPT-6 Astra with effort aliases (`-low` through `-ultra`); Codex CLI identity pins `@openai/codex@0.153.4` in lockstep with the image ([#13026](https://github.com/diegosouzapw/OmniRoute/pull/13026))
--- a/changelog.d/maintenance/layout-no-google-fonts.md
+++ b/changelog.d/maintenance/layout-no-google-fonts.md
@@ -0,0 +1 @@
+- **build:** root layout no longer loads Inter from `next/font/google`, so a production image build does not need fonts.googleapis.com ([#13026](https://github.com/diegosouzapw/OmniRoute/pull/13026))
--- a/docs/reference/PROVIDER_REFERENCE.md
+++ b/docs/reference/PROVIDER_REFERENCE.md
@@ -1,14 +1,14 @@
 ---
 title: "Provider Reference"
 version: 3.8.51
-lastUpdated: 2026-09-03
+lastUpdated: 2026-09-05
 ---

 # Provider Reference

 > **Auto-generated** from `src/shared/constants/providers.ts` — do not edit by hand.
 > Regenerate with: `npm run gen:provider-reference`
-> **Last generated:** 2026-09-03
+> **Last generated:** 2026-09-05

 Total providers: **356**. See category breakdown below.

--- a/open-sse/config/providers/registry/codex/index.ts
+++ b/open-sse/config/providers/registry/codex/index.ts
@@ -28,6 +28,25 @@ export const codexProvider: RegistryEntry = {
    tokenUrl: "https://auth.openai.com/oauth/token",
  },
  models: [
+    // Astra shares GPT-5.6's Codex limits: the live OAuth catalog reports
+    // max_context_window=872000 (context_window=272000 is the pricing tier).
+    { id: "gpt-6-astra", name: "GPT 6 Astra", ...GPT_5_6_CODEX_CAPABILITIES },
+    { id: "gpt-6-astra-ultra", name: "GPT 6 Astra (Ultra)", ...GPT_5_6_CODEX_CAPABILITIES },
+    { id: "gpt-6-astra-max", name: "GPT 6 Astra (Max)", ...GPT_5_6_CODEX_CAPABILITIES },
+    {
+      id: "gpt-6-astra-xhigh",
+      name: "GPT 6 Astra (xHigh)",
+      ...GPT_5_6_CODEX_CAPABILITIES,
+      timeoutMs: 1200000,
+    },
+    {
+      id: "gpt-6-astra-high",
+      name: "GPT 6 Astra (High)",
+      ...GPT_5_6_CODEX_CAPABILITIES,
+      timeoutMs: 1200000,
+    },
+    { id: "gpt-6-astra-medium", name: "GPT 6 Astra (Medium)", ...GPT_5_6_CODEX_CAPABILITIES },
+    { id: "gpt-6-astra-low", name: "GPT 6 Astra (Low)", ...GPT_5_6_CODEX_CAPABILITIES },
    {
      id: "gpt-5.6-sol",
      name: "GPT 5.6 Sol",
--- a/open-sse/config/providers/registry/openai/index.ts
+++ b/open-sse/config/providers/registry/openai/index.ts
@@ -12,6 +12,15 @@ export const openaiProvider: RegistryEntry = {
  authHeader: "bearer",
  defaultContextLength: 128000,
  models: [
+    // Astra shares the public GPT-5.6 limits; tool calling requires Responses.
+    // https://developers.openai.com/api/docs/guides/latest-model
+    {
+      id: "gpt-6-astra",
+      name: "GPT-6 Astra",
+      ...GPT_5_6_API_CAPABILITIES,
+      supportedThinkingEfforts: ["low", "medium", "high", "xhigh", "max"],
+      unsupportedParams: ["temperature", "top_p", "top_logprobs", "logprobs"],
+    },
    // #11489: per OpenAI's model reference `gpt-5.6` is an ALIAS of `gpt-5.6-sol`,
    // not a distinct model — quality scores point forward, which no suffix
    // stripper can express. Siblings `-terra`/`-luna` are their own models.
--- a/open-sse/executors/codex.ts
+++ b/open-sse/executors/codex.ts
@@ -54,7 +54,7 @@ export {
 import { isCodexFreePlan, normalizeCodexTools } from "./codex/tools.ts";
 import {
  CODEX_EFFORT_ORDER as EFFORT_ORDER,
-  GPT_5_6_ULTRA_ALIAS_MODELS,
+  CODEX_ULTRA_ALIAS_MODELS,
  splitCodexReasoningSuffix,
  type CodexEffortLevel as EffortLevel,
 } from "./codex/reasoningSuffix.ts";
@@ -167,13 +167,13 @@ function isCodexResponsesLiteRequest(
  );
 }

-// GPT-5.6 ultra-tier (sol/terra at "ultra") and luna at "max" coordinate delegation to
+// Astra/Sol/Terra at "ultra" and Luna at "max" coordinate delegation to
 // sub-agents via parallel tool calls (see the effort-clamp comment near clampEffort()).
 // Responses Lite must not strip parallel_tool_calls for those model/effort combos, or
 // delegation silently breaks while the request still returns HTTP 200 (issue #7821).
 function isCodexDelegationDependentModel(model: unknown): boolean {
  const { baseModel, effort } = splitCodexReasoningSuffix(model);
-  if (effort === "ultra" && GPT_5_6_ULTRA_ALIAS_MODELS.has(baseModel)) return true;
+  if (effort === "ultra" && CODEX_ULTRA_ALIAS_MODELS.has(baseModel)) return true;
  if (effort === "max" && baseModel === "gpt-5.6-luna") return true;
  return false;
 }
@@ -324,12 +324,9 @@ function normalizeServiceTierValue(value: unknown): string | undefined {
  return normalized;
 }

-/**
- * Maximum reasoning effort allowed per Codex model.
- * Models not listed here retain the legacy xhigh cap.
- * Update this table when Codex releases new models with different caps.
- */
+/** Maximum reasoning effort per Codex model; unlisted models keep the xhigh cap. */
 const MAX_EFFORT_BY_MODEL: Record<string, EffortLevel> = {
+  "gpt-6-astra": "ultra",
  "gpt-5.6-sol": "ultra",
  "gpt-5.6-terra": "ultra",
  "gpt-5.6-luna": "max",
--- a/open-sse/executors/codex/reasoningSuffix.ts
+++ b/open-sse/executors/codex/reasoningSuffix.ts
@@ -8,25 +8,28 @@ export const CODEX_EFFORT_ORDER = [
  "ultra",
 ] as const;
 export type CodexEffortLevel = (typeof CODEX_EFFORT_ORDER)[number];
-export const GPT_5_6_MAX_ALIAS_MODELS = new Set(["gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"]);
-export const GPT_5_6_ULTRA_ALIAS_MODELS = new Set(["gpt-5.6-sol", "gpt-5.6-terra"]);
+export const CODEX_MAX_ALIAS_MODELS = new Set([
+  "gpt-5.6-sol",
+  "gpt-5.6-terra",
+  "gpt-5.6-luna",
+  "gpt-6-astra",
+]);
+export const CODEX_ULTRA_ALIAS_MODELS = new Set(["gpt-5.6-sol", "gpt-5.6-terra", "gpt-6-astra"]);

 export function splitCodexReasoningSuffix(model: unknown): {
  baseModel: string;
  effort: CodexEffortLevel | null;
 } {
  const modelId = typeof model === "string" ? model : "";
-  const gpt56Match = /^(gpt-5\.6-(?:sol|terra|luna))(?:-(max|ultra)|\((max|ultra)\))$/.exec(
-    modelId
-  );
-  if (gpt56Match) {
-    const [, baseModel, hyphenEffort, parenthesizedEffort] = gpt56Match;
+  const maxTierMatch = /^(.+?)(?:-(max|ultra)|\((max|ultra)\))$/.exec(modelId);
+  if (maxTierMatch) {
+    const [, baseModel, hyphenEffort, parenthesizedEffort] = maxTierMatch;
    const effort = hyphenEffort ?? parenthesizedEffort;
    const supportedModels = parenthesizedEffort
-      ? GPT_5_6_MAX_ALIAS_MODELS
+      ? CODEX_MAX_ALIAS_MODELS
      : effort === "ultra"
-        ? GPT_5_6_ULTRA_ALIAS_MODELS
-        : GPT_5_6_MAX_ALIAS_MODELS;
+        ? CODEX_ULTRA_ALIAS_MODELS
+        : CODEX_MAX_ALIAS_MODELS;
    if (supportedModels.has(baseModel)) {
      return { baseModel, effort: effort as CodexEffortLevel };
    }
--- a/open-sse/translator/request/openai-responses/helpers.ts
+++ b/open-sse/translator/request/openai-responses/helpers.ts
@@ -50,8 +50,8 @@ export function imageUrlToText(value: unknown): string {
  return toString(record.url);
 }

-const CODEX_GPT_5_6_MODEL_PATTERN =
-  /^gpt-5\.6-(?:sol|terra|luna)(?:-(?:none|low|medium|high|xhigh|max|ultra))?$/;
+const CODEX_MAX_EFFORT_MODEL_PATTERN =
+  /^(?:gpt-5\.6-(?:sol|terra|luna)|gpt-6-astra)(?:-(?:none|low|medium|high|xhigh|max|ultra))?$/;
 const KIRO_GPT_5_6_MODEL_PATTERN =
  /^(?:kiro|kr)\/gpt-5\.6-(?:sol|terra|luna)(?:-(?:none|low|medium|high|xhigh|max))?$/;

@@ -61,7 +61,7 @@ function supportsNativeMaxReasoningEffort(model: unknown): boolean {
    .toLowerCase()
    .replace(/^(?:codex|cx)\//, "");
  return (
-    CODEX_GPT_5_6_MODEL_PATTERN.test(normalizedModel) ||
+    CODEX_MAX_EFFORT_MODEL_PATTERN.test(normalizedModel) ||
    KIRO_GPT_5_6_MODEL_PATTERN.test(toString(model).trim().toLowerCase())
  );
 }
--- a/src/app/layout.tsx
+++ b/src/app/layout.tsx
@@ -1,4 +1,3 @@
-import { Inter } from "next/font/google";
 import "./globals.css";
 import { ThemeProvider } from "@/shared/components/ThemeProvider";
 import { NextIntlClientProvider } from "next-intl";
@@ -11,11 +10,6 @@ import { PwaRegister } from "@/shared/components/PwaRegister";
 import { LocaleAutoDetect } from "@/shared/components/LocaleAutoDetect";
 import { BasePathNetworkProvider } from "@/shared/components/BasePathNetworkProvider";

-const inter = Inter({
-  subsets: ["latin"],
-  variable: "--font-inter",
-});
-
 export const viewport: Viewport = {
  themeColor: "#0b0f1a",
  viewportFit: "cover",
@@ -135,7 +129,7 @@ export default async function RootLayout({ children }) {
          }}
        />
      </head>
-      <body className={`${inter.variable} font-sans antialiased`} suppressHydrationWarning>
+      <body className="font-sans antialiased" suppressHydrationWarning>
        <a
          href="#main-content"
          className="sr-only focus:not-sr-only focus:absolute focus:top-2 focus:left-2 focus:z-50 focus:px-4 focus:py-2 focus:bg-[#6366f1] focus:text-white focus:rounded-lg focus:text-sm focus:font-semibold focus:shadow-lg"
--- a/src/lib/usage/costCalculator.ts
+++ b/src/lib/usage/costCalculator.ts
@@ -101,6 +101,8 @@ export function getCodexFastCostMultiplier(

  const modelKey = stripCodexEffortSuffix(normalizeModelName(String(model || "")).toLowerCase());
  const compactModelKey = modelKey.replace(/-/g, "");
+  // Codex Astra Fast is 2.5x Standard (https://developers.openai.com/codex/pricing).
+  if (modelKey === "gpt-6-astra" || compactModelKey === "gpt6astra") return 2.5;
  if (
    /^gpt-5\.6-(?:sol|terra|luna)$/.test(modelKey) ||
    /^gpt5\.6(?:sol|terra|luna)$/.test(compactModelKey)
--- a/src/shared/constants/codexClient.ts
+++ b/src/shared/constants/codexClient.ts
@@ -3,7 +3,7 @@
 // refresh this so the fingerprint OpenAI sees from the OAuth/Responses face
 // matches the real client version. Overridable per-deployment via
 // CODEX_CLIENT_VERSION.
-export const DEFAULT_CODEX_CLIENT_VERSION = "0.153.2";
+export const DEFAULT_CODEX_CLIENT_VERSION = "0.153.4";
 export const CODEX_CLI_RS_ORIGINATOR = "codex_cli_rs";

 export function getCodexCliRsHeaders(
--- a/src/shared/constants/modelSpecs.ts
+++ b/src/shared/constants/modelSpecs.ts
@@ -118,6 +118,12 @@ const GEMINI_36_FLASH_MODEL_SPEC = {
 } satisfies ModelSpec;

 export const MODEL_SPECS: Record<string, ModelSpec> = {
+  // Public model limits; the Codex registry supplies its smaller OAuth window.
+  // https://developers.openai.com/api/docs/models/gpt-6-astra
+  "gpt-6-astra": {
+    ...GPT_5_6_MODEL_SPEC,
+    aliases: ["openai/gpt-6-astra"],
+  },
  "gpt-5.6": {
    ...GPT_5_6_MODEL_SPEC,
    aliases: ["openai/gpt-5.6"],
--- a/src/shared/constants/pricing/frontier-labs.ts
+++ b/src/shared/constants/pricing/frontier-labs.ts
@@ -3,6 +3,7 @@
 * Pure data; merged by default-pricing.ts via spread (god-file decomposition; semantic split).
 */
 import {
+  GPT_6_ASTRA_PRICING,
  GEMINI_3_7_FLASH_PROMO_PRICING,
  GPT_5_5_PRICING,
  GPT_5_6_LUNA_PRICING,
@@ -20,6 +21,7 @@ import {

 export const DEFAULT_PRICING_FRONTIER = {
  openai: {
+    "gpt-6-astra": GPT_6_ASTRA_PRICING,
    "gpt-5.6": GPT_5_6_SOL_PRICING,
    "gpt-5.6-sol": GPT_5_6_SOL_PRICING,
    "gpt-5.6-terra": GPT_5_6_TERRA_PRICING,
--- a/src/shared/constants/pricing/oauth-subscriptions.ts
+++ b/src/shared/constants/pricing/oauth-subscriptions.ts
@@ -3,6 +3,7 @@
 * Pure data; merged by default-pricing.ts via spread (god-file decomposition; semantic split).
 */
 import {
+  GPT_6_ASTRA_PRICING,
  CLAUDE_FABLE_5_1_PRICING,
  CLAUDE_OPUS_5_PRICING,
  GEMINI_3_7_FLASH_PROMO_PRICING,
@@ -19,6 +20,10 @@ const ANTIGRAVITY_GEMINI_3_7_PRICING = {
  "gemini-3.7-flash-high": GEMINI_3_7_FLASH_PROMO_PRICING,
 };

+// Codex Standard: 250 / 25 / 1250 credits per MTok, at 25 credits per USD.
+// https://developers.openai.com/codex/pricing
+const GPT_6_ASTRA_CODEX_PRICING = GPT_6_ASTRA_PRICING;
+
 export const DEFAULT_PRICING_OAUTH = {
  cc: {
    "claude-fable-5-1": CLAUDE_FABLE_5_1_PRICING,
@@ -88,6 +93,13 @@ export const DEFAULT_PRICING_OAUTH = {
    },
  },
  cx: {
+    "gpt-6-astra": GPT_6_ASTRA_CODEX_PRICING,
+    "gpt-6-astra-ultra": GPT_6_ASTRA_CODEX_PRICING,
+    "gpt-6-astra-max": GPT_6_ASTRA_CODEX_PRICING,
+    "gpt-6-astra-xhigh": GPT_6_ASTRA_CODEX_PRICING,
+    "gpt-6-astra-high": GPT_6_ASTRA_CODEX_PRICING,
+    "gpt-6-astra-medium": GPT_6_ASTRA_CODEX_PRICING,
+    "gpt-6-astra-low": GPT_6_ASTRA_CODEX_PRICING,
    "codex-auto-review": GPT_5_5_PRICING,
    // Codex uses credits per 1M tokens. OmniRoute stores the dollar-equivalent
    // values below at the documented conversion of 25 credits per USD.
--- a/src/shared/constants/pricing/shared-tiers.ts
+++ b/src/shared/constants/pricing/shared-tiers.ts
@@ -1,6 +1,16 @@
 /**
 * Pricing data — shared per-MTok tier constants (god-file decomposition). Pure data; merged by the barrel.
 */
+// OpenAI API Standard; Codex Standard has the same dollar-equivalent rates.
+// https://openai.com/index/gpt-6-astra/
+export const GPT_6_ASTRA_PRICING = {
+  input: 10.0,
+  output: 50.0,
+  cached: 1.0,
+  reasoning: 50.0,
+  cache_creation: 12.5,
+};
+
 export const GPT_5_3_CODEX_PRICING = {
  input: 5.0,
  output: 20.0,
--- a/tests/snapshots/provider/translate-path.json
+++ b/tests/snapshots/provider/translate-path.json
@@ -1311,16 +1311,16 @@
        "Authorization": "Bearer <TOK>",
        "Content-Type": "application/json",
        "Openai-Beta": "responses=experimental",
-        "User-Agent": "codex-cli/0.153.2 (<OS>; <ARCH>)",
-        "Version": "0.153.2",
+        "User-Agent": "codex-cli/0.153.4 (<OS>; <ARCH>)",
+        "Version": "0.153.4",
        "X-Codex-Beta-Features": "responses_websockets"
      },
      "nonStream": {
        "Authorization": "Bearer <TOK>",
        "Content-Type": "application/json",
        "Openai-Beta": "responses=experimental",
-        "User-Agent": "codex-cli/0.153.2 (<OS>; <ARCH>)",
-        "Version": "0.153.2",
+        "User-Agent": "codex-cli/0.153.4 (<OS>; <ARCH>)",
+        "Version": "0.153.4",
        "X-Codex-Beta-Features": "responses_websockets"
      },
      "oauth": {
@@ -1328,8 +1328,8 @@
        "Authorization": "Bearer <TOK>",
        "Content-Type": "application/json",
        "Openai-Beta": "responses=experimental",
-        "User-Agent": "codex-cli/0.153.2 (<OS>; <ARCH>)",
-        "Version": "0.153.2",
+        "User-Agent": "codex-cli/0.153.4 (<OS>; <ARCH>)",
+        "Version": "0.153.4",
        "X-Codex-Beta-Features": "responses_websockets"
      }
    },
--- a/tests/unit/agentrouter-chatcore-protocols.test.ts
+++ b/tests/unit/agentrouter-chatcore-protocols.test.ts
@@ -112,7 +112,7 @@ test("AgentRouter Responses requests automatically use the native Responses prot
      body: structuredClone(body),
      headers: new Headers({ accept: "application/json", originator: "codex_cli_rs" }),
    },
-    userAgent: "codex_cli_rs/0.149.0",
+    userAgent: "codex_cli_rs/0.153.4",
  });

  assert.ok(captured);
@@ -176,7 +176,7 @@ test("AgentRouter OpenAI Chat requests automatically use the native Chat protoco
      body: structuredClone(body),
      headers: new Headers({ accept: "application/json" }),
    },
-    userAgent: "codex_cli_rs/0.149.0",
+    userAgent: "codex_cli_rs/0.153.4",
  });

  assert.equal(result.success, true);
@@ -304,7 +304,7 @@ test("AgentRouter Responses streaming stays native without a connection protocol
      body: structuredClone(body),
      headers: new Headers({ accept: "text/event-stream", originator: "codex_cli_rs" }),
    },
-    userAgent: "codex_cli_rs/0.149.0",
+    userAgent: "codex_cli_rs/0.153.4",
  });

  assert.equal(result.success, true);
@@ -383,7 +383,7 @@ test("AgentRouter OpenAI Chat streaming stays native without a connection protoc
      body: structuredClone(body),
      headers: new Headers({ accept: "text/event-stream" }),
    },
-    userAgent: "codex_cli_rs/0.149.0",
+    userAgent: "codex_cli_rs/0.153.4",
  });

  assert.equal(result.success, true);
--- a/tests/unit/claude-codex-identity-version-sync.test.ts
+++ b/tests/unit/claude-codex-identity-version-sync.test.ts
@@ -127,3 +127,39 @@ test("test 7: live-empty GitHub catalog path does not call persist", () => {
  const liveWindow = src.slice(liveIdx, start);
  assert.match(liveWindow, /buildApiDiscoveryResponse\s*\(/);
 });
+
+test("Codex client version locksteps Dockerfile @openai/codex and env override", () => {
+  const dockerfile = fs.readFileSync(path.join(process.cwd(), "Dockerfile"), "utf8");
+  const match = dockerfile.match(/@openai\/codex@([0-9]+\.[0-9]+\.[0-9]+)/);
+  assert.ok(match, "Dockerfile must pin @openai/codex@x.y.z");
+  const pinned = match[1];
+  assert.notEqual(pinned, "0.149.0");
+  assert.equal(codexCfg.DEFAULT_CODEX_CLIENT_VERSION, pinned);
+  assert.equal(codexCfg.getCodexClientVersion(), pinned);
+  assert.equal(codexCfg.getCodexDefaultHeaders().Version, pinned);
+  assert.equal(
+    codexCfg.getCodexCliRsHeaders()["User-Agent"],
+    `codex_cli_rs/${pinned}`,
+  );
+});
+
+test("test 7: live-empty GitHub catalog path does not call persist", () => {
+  const src = fs.readFileSync(
+    path.join(process.cwd(), "src/app/api/providers/[id]/models/route.ts"),
+    "utf8",
+  );
+  // The githubCatalogModels fallback must use buildResponse, not buildApiDiscoveryResponse.
+  const idx = src.indexOf("Codex live catalog unavailable — using GitHub model catalog");
+  assert.ok(idx > 0);
+  const start = src.lastIndexOf("if (githubCatalogModels", idx);
+  const end = src.indexOf("if (cachedDiscoveryModels", idx);
+  assert.ok(start > 0 && end > start);
+  const window = src.slice(start, end);
+  assert.match(window, /buildResponse\s*\(/);
+  assert.doesNotMatch(window, /buildApiDiscoveryResponse\s*\(/);
+
+  const liveIdx = src.lastIndexOf("if (liveModels && liveModels.length > 0)");
+  assert.ok(liveIdx > 0 && liveIdx < start);
+  const liveWindow = src.slice(liveIdx, start);
+  assert.match(liveWindow, /buildApiDiscoveryResponse\s*\(/);
+});
--- a/tests/unit/codex-astra.test.ts
+++ b/tests/unit/codex-astra.test.ts
@@ -0,0 +1,109 @@
+import test from "node:test";
+import assert from "node:assert/strict";
+
+import { getModelsByProviderId } from "../../open-sse/config/providerModels.ts";
+import { CodexExecutor } from "../../open-sse/executors/codex.ts";
+import { openaiToOpenAIResponsesRequest } from "../../open-sse/translator/request/openai-responses/toResponses.ts";
+import { getModelSpec } from "../../src/shared/constants/modelSpecs.ts";
+import { getPricingForModel } from "../../src/shared/constants/pricing.ts";
+import { getCodexFastCostMultiplier } from "../../src/lib/usage/costCalculator.ts";
+
+const MODEL = "gpt-6-astra";
+const EFFORTS = ["ultra", "max", "xhigh", "high", "medium", "low"] as const;
+
+test.after(async () => {
+  const { resetDbInstance } = await import("../../src/lib/db/core.ts");
+  resetDbInstance();
+});
+
+test("Codex exposes Astra and its effort variants with live OAuth limits", () => {
+  const ids = [MODEL, ...EFFORTS.map((effort) => `${MODEL}-${effort}`)];
+  for (const provider of ["codex", "codex-app-server"]) {
+    const models = getModelsByProviderId(provider);
+    assert.deepEqual(
+      models.filter((model) => model.id.startsWith(MODEL)).map((model) => model.id),
+      ids
+    );
+    for (const id of ids) {
+      const model = models.find((entry) => entry.id === id);
+      assert.ok(model, `${provider}/${id}`);
+      assert.equal(model.contextLength, 872000);
+      assert.equal(model.maxInputTokens, 872000);
+      assert.equal(model.maxOutputTokens, 128000);
+      assert.equal(model.targetFormat, "openai-responses");
+      assert.equal(model.toolCalling, true);
+      assert.equal(model.supportsReasoning, true);
+      assert.equal(model.supportsVision, true);
+      assert.equal(model.supportsXHighEffort, true);
+    }
+    assert.deepEqual(
+      models.slice(0, ids.length).map((model) => model.id),
+      ids
+    );
+  }
+});
+
+test("Astra specs retain the public context window separately from Codex limits", () => {
+  const spec = getModelSpec(MODEL);
+  assert.equal(spec?.contextWindow, 1050000);
+  assert.equal(spec?.maxOutputTokens, 128000);
+  assert.equal(spec?.supportsTools, true);
+  assert.equal(spec?.supportsVision, true);
+  assert.equal(spec?.supportsThinking, true);
+});
+
+test("Astra effort aliases reach Codex as the base model and supported wire effort", () => {
+  const executor = new CodexExecutor();
+  for (const effort of EFFORTS) {
+    const model = `${MODEL}-${effort}`;
+    const result = executor.transformRequest(model, { model, input: [] }, false, {
+      requestEndpointPath: "/responses",
+    });
+    assert.equal(result.model, MODEL, effort);
+    assert.equal(result.reasoning.effort, effort === "ultra" ? "max" : effort, effort);
+  }
+});
+
+test("Chat-to-Codex translation preserves Astra max reasoning", () => {
+  const translated = openaiToOpenAIResponsesRequest(
+    MODEL,
+    { model: MODEL, messages: [{ role: "user", content: "test" }], reasoning_effort: "max" },
+    true,
+    {}
+  );
+  const result = new CodexExecutor().transformRequest(MODEL, translated, true, {
+    requestEndpointPath: "/chat/completions",
+  });
+  assert.equal(result.model, MODEL);
+  assert.equal(result.reasoning.effort, "max");
+});
+
+test("Astra parenthesized effort overrides preserve the reasoning summary", () => {
+  for (const effort of ["max", "ultra"]) {
+    const model = `${MODEL}(${effort})`;
+    const result = new CodexExecutor().transformRequest(
+      model,
+      { model, input: [], reasoning: { effort: "low", summary: "detailed" } },
+      false,
+      { requestEndpointPath: "/responses" }
+    );
+    assert.equal(result.model, MODEL);
+    assert.equal(result.reasoning.effort, "max");
+    assert.equal(result.reasoning.summary, "detailed");
+  }
+});
+
+test("Astra Codex pricing and Fast multiplier match the Codex credit rate card", () => {
+  for (const model of [MODEL, ...EFFORTS.map((effort) => `${MODEL}-${effort}`)]) {
+    const pricing = getPricingForModel("cx", model);
+    assert.ok(pricing, model);
+    assert.equal(pricing.input, 10);
+    assert.equal(pricing.cached, 1);
+    assert.equal(pricing.output, 50);
+    assert.equal(pricing.reasoning, 50);
+    assert.equal(getCodexFastCostMultiplier("codex", model, "priority"), 2.5);
+    assert.equal(getCodexFastCostMultiplier("cx", model, "fast"), 2.5);
+    assert.equal(getCodexFastCostMultiplier("codex", model, "default"), 1);
+  }
+  assert.equal(getCodexFastCostMultiplier("openai", MODEL, "priority"), 1);
+});
--- a/tests/unit/codex-gpt56-catalog.test.ts
+++ b/tests/unit/codex-gpt56-catalog.test.ts
@@ -29,7 +29,7 @@ test("Codex catalog exposes the GPT-5.6 lineup in configured priority order", ()
  ];

  assert.deepEqual(
-    models.slice(0, expectedIds.length).map((model) => model.id),
+    models.filter((model) => model.id.startsWith("gpt-5.6-")).map((model) => model.id),
    expectedIds
  );

--- a/tests/unit/executor-codex-gpt56-lite-ultra.test.ts
+++ b/tests/unit/executor-codex-gpt56-lite-ultra.test.ts
@@ -70,6 +70,12 @@ test("Responses Lite must not strip parallel_tool_calls for GPT-5.6 luna max-tie
  assert.equal(capturedBodies[0].parallel_tool_calls, true);
 });

+test("Responses Lite preserves parallel tool calls for Astra ultra delegation", async () => {
+  const capturedBodies = await runLiteRequest("gpt-6-astra-ultra");
+  assert.equal(capturedBodies.length, 1);
+  assert.equal(capturedBodies[0].parallel_tool_calls, true);
+});
+
 test("Responses Lite still forces parallel_tool_calls:false for non-delegation GPT-5.5", async () => {
  const capturedBodies = await runLiteRequest("gpt-5.5");
  assert.equal(
--- a/tests/unit/executor-codex.test.ts
+++ b/tests/unit/executor-codex.test.ts
@@ -184,10 +184,10 @@ test("CodexExecutor.buildHeaders binds workspace ids and disables SSE accept for
  assert.equal(standardHeaders.Authorization, "Bearer codex-token");
  assert.equal(standardHeaders.Accept, "text/event-stream");
  assert.equal(standardHeaders["chatgpt-account-id"], "workspace-1");
-  assert.equal(standardHeaders.Version, "0.153.2");
+  assert.equal(standardHeaders.Version, "0.153.4");
  assert.equal(standardHeaders["Openai-Beta"], "responses=experimental");
  assert.equal(standardHeaders["X-Codex-Beta-Features"], "responses_websockets");
-  assert.equal(standardHeaders["User-Agent"], "codex-cli/0.153.2 (Windows 10.0.26200; x64)");
+  assert.equal(standardHeaders["User-Agent"], "codex-cli/0.153.4 (Windows 10.0.26200; x64)");
  assert.equal(compactHeaders.Accept, "application/json");
 });

@@ -213,7 +213,7 @@ test("CodexExecutor.buildHeaders honors safe env overrides for Version and User-
    },
    () => {
      const headers = executor.buildHeaders({ accessToken: "codex-token" }, true);
-      assert.equal(headers.Version, "0.153.2");
+      assert.equal(headers.Version, "0.153.4");
      assert.equal(headers["User-Agent"], "custom-codex/9.9.9");
    }
  );
--- a/tests/unit/openai-gpt56-catalog.test.ts
+++ b/tests/unit/openai-gpt56-catalog.test.ts
@@ -4,10 +4,14 @@ import assert from "node:assert/strict";
 import { getModelsByProviderId } from "../../open-sse/config/providerModels.ts";
 import { getModelSpec } from "../../src/shared/constants/modelSpecs.ts";
 import { getPricingForModel } from "../../src/shared/constants/pricing.ts";
+import { getUnsupportedParams } from "../../open-sse/config/providerRegistry.ts";
+import { DefaultExecutor } from "../../open-sse/executors/default.ts";
+import { resolveChatCoreTargetFormat } from "../../open-sse/handlers/chatCore/targetFormat.ts";
+import { openaiToOpenAIResponsesRequest } from "../../open-sse/translator/request/openai-responses/toResponses.ts";

-const EXPECTED_MODELS = ["gpt-5.6", "gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"];
+const EXPECTED_MODELS = ["gpt-6-astra", "gpt-5.6", "gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"];

-test("OpenAI API catalog exposes the public GPT-5.6 family and keeps GPT-5.4", () => {
+test("OpenAI API catalog puts Astra before GPT-5.6 and keeps GPT-5.4", () => {
  const models = getModelsByProviderId("openai");

  assert.deepEqual(
@@ -38,8 +42,9 @@ test("OpenAI API catalog exposes the public GPT-5.6 family and keeps GPT-5.4", (
  }
 });

-test("OpenAI API GPT-5.6 pricing matches the published standard tier", () => {
+test("OpenAI API Astra and GPT-5.6 pricing matches the published standard tier", () => {
  const expectedPricing = {
+    "gpt-6-astra": { input: 10, cached: 1, cache_creation: 12.5, output: 50 },
    "gpt-5.6": { input: 5, cached: 0.5, cache_creation: 6.25, output: 30 },
    "gpt-5.6-sol": { input: 5, cached: 0.5, cache_creation: 6.25, output: 30 },
    "gpt-5.6-terra": { input: 2.5, cached: 0.25, cache_creation: 3.125, output: 15 },
@@ -55,3 +60,72 @@ test("OpenAI API GPT-5.6 pricing matches the published standard tier", () => {
    assert.equal(pricing.output, expected.output, `${modelId} output`);
  }
 });
+
+test("OpenAI Astra declares unsupported sampling parameters for the chat pipeline", () => {
+  assert.deepEqual(getUnsupportedParams("openai", "gpt-6-astra"), [
+    "temperature",
+    "top_p",
+    "top_logprobs",
+    "logprobs",
+  ]);
+});
+
+test("OpenAI Astra tool requests use Responses and retain each supported reasoning effort", async () => {
+  const model = "gpt-6-astra";
+  assert.equal(
+    resolveChatCoreTargetFormat({
+      provider: "openai",
+      resolvedModel: model,
+      apiFormat: undefined,
+      customModelTargetFormat: undefined,
+      providerSpecificData: null,
+    }).targetFormat,
+    "openai-responses"
+  );
+  const originalFetch = globalThis.fetch;
+  const captured: Array<{ url: string; body: Record<string, unknown> }> = [];
+  globalThis.fetch = async (url, init) => {
+    captured.push({ url: String(url), body: JSON.parse(String(init?.body || "{}")) });
+    return new Response(JSON.stringify({ id: "resp_astra", object: "response", output: [] }), {
+      status: 200,
+      headers: { "Content-Type": "application/json" },
+    });
+  };
+  try {
+    for (const effort of ["low", "medium", "high", "xhigh", "max", "none", "minimal"]) {
+      const body = openaiToOpenAIResponsesRequest(
+        model,
+        {
+          model,
+          messages: [{ role: "user", content: "Call the test tool." }],
+          reasoning_effort: effort,
+          tools: [
+            {
+              type: "function",
+              function: { name: "test_tool", parameters: { type: "object", properties: {} } },
+            },
+          ],
+        },
+        false,
+        {}
+      );
+      await new DefaultExecutor("openai").execute({
+        model,
+        body,
+        stream: false,
+        credentials: { apiKey: "test-openai-key" },
+      });
+      const request = captured.at(-1)!;
+      assert.equal(request.url, "https://api.openai.com/v1/responses");
+      assert.equal(request.body.model, model);
+      assert.equal(
+        (request.body.reasoning as { effort: string }).effort,
+        effort === "none" || effort === "minimal" ? "low" : effort
+      );
+      assert.ok(Array.isArray(request.body.input));
+      assert.equal((request.body.tools as Array<{ name: string }>)[0].name, "test_tool");
+    }
+  } finally {
+    globalThis.fetch = originalFetch;
+  }
+});
				`@@ -0,0 +1 @@`
				- feat(sse): Codex and OpenAI catalogs list GPT-6 Astra with effort aliases (`-low` through `-ultra`); Codex CLI identity pins `@openai/codex@0.153.4` in lockstep with the image ([#13026](https://github.com/diegosouzapw/OmniRoute/pull/13026))
				`@@ -0,0 +1 @@`
				- build: root layout no longer loads Inter from `next/font/google`, so a production image build does not need fonts.googleapis.com ([#13026](https://github.com/diegosouzapw/OmniRoute/pull/13026))