@cartanova/qgrid-cli 2.4.6 → 2.4.8

package/bundle/web-dist/client/index.html

@@ -5,12 +5,12 @@
     <link rel="icon" type="image/svg+xml" href="/sonamu.svg" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
     <title>Qgrid</title>
-    <script type="module" crossorigin src="/assets/index-BOXhNo0k.js"></script>
+    <script type="module" crossorigin src="/assets/index-CmuPjbCl.js"></script>
     <link rel="modulepreload" crossorigin href="/assets/rolldown-runtime-Dw2cE7zH.js">
     <link rel="modulepreload" crossorigin href="/assets/preload-helper-ca-nBW7U.js">
     <link rel="modulepreload" crossorigin href="/assets/vendor-react-DH4kUfK2.js">
     <link rel="modulepreload" crossorigin href="/assets/vendor-tanstack-DJRQAbrY.js">
-    <link rel="modulepreload" crossorigin href="/assets/sd.generated-D2Y9QeiN.js">
+    <link rel="modulepreload" crossorigin href="/assets/sd.generated-DYuKzBY6.js">
     <link rel="stylesheet" crossorigin href="/assets/index-yZcjDoV0.css">
   </head>
   <body>

package/bundle/web-dist/server/assets/{lazyRouteComponent-BlfOh9S3.js → lazyRouteComponent-Cu6IkL8A.js}

@@ -1,4 +1,4 @@
-import { J as __commonJSMin, Z as __toESM, _ as require_jsx_runtime, m as require_shim, p as require_react_dom, v as require_react } from "./sd.generated-CE6D5clm.js";
+import { J as __commonJSMin, Z as __toESM, _ as require_jsx_runtime, m as require_shim, p as require_react_dom, v as require_react } from "./sd.generated-qzuRe35X.js";
 //#region ../../node_modules/.pnpm/@tanstack+router-core@1.143.6/node_modules/@tanstack/router-core/dist/esm/utils.js
 function last(arr) {
 	return arr[arr.length - 1];

package/bundle/web-dist/server/assets/{logs-DYly4ipQ.js → logs-Dx2knN2T.js}

@@ -1,6 +1,6 @@
-import { _ as require_jsx_runtime, d as string, l as number, u as object, v as require_react } from "./sd.generated-CE6D5clm.js";
-import { a as TokenService, n as QgridService, r as RequestLogService } from "./services.generated-DyZUzXiq.js";
-import { a as Link, o as useNavigate } from "./lazyRouteComponent-BlfOh9S3.js";
+import { _ as require_jsx_runtime, d as string, l as number, u as object, v as require_react } from "./sd.generated-qzuRe35X.js";
+import { a as TokenService, n as QgridService, r as RequestLogService } from "./services.generated-C8q1M41a.js";
+import { a as Link, o as useNavigate } from "./lazyRouteComponent-Cu6IkL8A.js";
 import { n as Route } from "../entry-server.generated.js";
 import { n as formatMicroUsd, r as formatUsd, t as cacheHitRate } from "./cost-CSH_t17I.js";
 require_react();
@@ -271,12 +271,15 @@ function RequestLogTable({ search, onSearchChange }) {
 									children: row.token_name
 								})
 							}),
-							/* @__PURE__ */ (0, import_jsx_runtime.jsx)("td", {
+							/* @__PURE__ */ (0, import_jsx_runtime.jsxs)("td", {
 								className: "px-4 py-1.5 whitespace-nowrap",
-								children: /* @__PURE__ */ (0, import_jsx_runtime.jsx)("span", {
+								children: [/* @__PURE__ */ (0, import_jsx_runtime.jsx)("span", {
 									className: "text-xs text-sand-500",
-									children: row.model_name ?? "—"
-								})
+									children: row.requested_model_name && row.model_name && row.requested_model_name !== row.model_name ? `${row.requested_model_name} → ${row.model_name}` : row.model_name ?? "—"
+								}), (row.fallback_count ?? 0) > 0 && /* @__PURE__ */ (0, import_jsx_runtime.jsxs)("span", {
+									className: "ml-1 text-[9px] text-caution-500",
+									children: ["×", row.fallback_count]
+								})]
 							}),
 							/* @__PURE__ */ (0, import_jsx_runtime.jsx)("td", {
 								className: "px-4 py-1.5 text-left tabular-nums text-sand-500 whitespace-nowrap",

package/bundle/web-dist/server/assets/{monit-CYyY2S2f.js → monit-DU6PLatf.js}

@@ -1,5 +1,5 @@
-import { Z as __toESM, _ as require_jsx_runtime, c as _enum, p as require_react_dom, u as object, v as require_react } from "./sd.generated-CE6D5clm.js";
-import { s as useQuery, t as MonitService } from "./services.generated-DyZUzXiq.js";
+import { Z as __toESM, _ as require_jsx_runtime, c as _enum, p as require_react_dom, u as object, v as require_react } from "./sd.generated-qzuRe35X.js";
+import { s as useQuery, t as MonitService } from "./services.generated-C8q1M41a.js";
 import { a as clsx, r as Route } from "../entry-server.generated.js";
 //#region ../../node_modules/.pnpm/@tanstack+virtual-core@3.17.4/node_modules/@tanstack/virtual-core/dist/esm/lazy-measurements.js
 var import_react = /* @__PURE__ */ __toESM(require_react(), 1);

package/bundle/web-dist/server/assets/{routes-Dqu3osXt.js → routes-CTco1oJZ.js}

@@ -1,8 +1,8 @@
-import { Z as __toESM, _ as require_jsx_runtime, g as useQueryClient, p as require_react_dom, v as require_react } from "./sd.generated-CE6D5clm.js";
-import { a as TokenService, n as QgridService } from "./services.generated-DyZUzXiq.js";
+import { Z as __toESM, _ as require_jsx_runtime, g as useQueryClient, p as require_react_dom, v as require_react } from "./sd.generated-qzuRe35X.js";
+import { a as TokenService, n as QgridService } from "./services.generated-C8q1M41a.js";
 import { i as Q } from "../entry-server.generated.js";
 import { r as formatUsd } from "./cost-CSH_t17I.js";
-import { t as useUpdateTokenMutation } from "./use-update-token-mutation-BejUW5sw.js";
+import { t as useUpdateTokenMutation } from "./use-update-token-mutation-BlZo6swF.js";
 //#region ../../node_modules/.pnpm/@dnd-kit+utilities@3.2.2_react@19.2.4/node_modules/@dnd-kit/utilities/dist/utilities.esm.js
 var import_react_dom = /* @__PURE__ */ __toESM(require_react_dom());
 var import_react = /* @__PURE__ */ __toESM(require_react());

package/bundle/web-dist/server/assets/{sd.generated-CE6D5clm.js → sd.generated-qzuRe35X.js}

@@ -22539,6 +22539,8 @@ var entityLabels = {
 	"entity.RequestLog.token_name": "토큰 이름",
 	"entity.RequestLog.project_name": "프로젝트 이름",
 	"entity.RequestLog.model_name": "모델 이름",
+	"entity.RequestLog.requested_model_name": "요청 모델 이름 (fallback 전)",
+	"entity.RequestLog.fallback_count": "모델 fallback 횟수",
 	"entity.RequestLog.user_prompt": "사용자 프롬프트",
 	"entity.RequestLog.system_prompt": "시스템 프롬프트",
 	"entity.RequestLog.response": "llm 응답 전문",
@@ -22546,9 +22548,12 @@ var entityLabels = {
 	"entity.RequestLog.output_tokens": "출력 토큰",
 	"entity.RequestLog.cache_read_tokens": "캐시 읽기 토큰",
 	"entity.RequestLog.cache_creation_tokens": "캐시 생성 토큰",
+	"entity.RequestLog.cache_creation_5m_tokens": "5분 TTL 캐시 생성 토큰",
+	"entity.RequestLog.cache_creation_1h_tokens": "1시간 TTL 캐시 생성 토큰",
 	"entity.RequestLog.duration_ms": "응답 소요시간(ms)",
 	"entity.RequestLog.ttft_ms": "first gen TTFT(ms)",
 	"entity.RequestLog.cost_usd": "API 비용",
+	"entity.RequestLog.cost_source": "비용 출처(provider/pricing_table/mixed)",
 	"entity.RequestLog.image_cost_usd": "이미지 생성 추정 비용",
 	"entity.RequestLog.image_cost_method": "이미지 비용 추정 방식",
 	"entity.RequestLog.effort": "effort level",
@@ -22571,10 +22576,17 @@ var entityLabels = {
 	"entity.RequestLogStep.request_log": "부모 request log ID",
 	"entity.RequestLogStep.step_index": "ai-sdk step number (0-based)",
 	"entity.RequestLogStep.type": "step 타입",
+	"entity.RequestLogStep.model_name": "실제 serving 모델",
+	"entity.RequestLogStep.requested_model_name": "요청 모델 (fallback 전)",
+	"entity.RequestLogStep.fallback_count": "모델 fallback 횟수",
 	"entity.RequestLogStep.input_tokens": "입력 토큰 (generate)",
 	"entity.RequestLogStep.output_tokens": "출력 토큰 (generate)",
 	"entity.RequestLogStep.cache_read_tokens": "캐시 읽기 토큰",
 	"entity.RequestLogStep.cache_creation_tokens": "캐시 생성 토큰",
+	"entity.RequestLogStep.cache_creation_5m_tokens": "5분 TTL 캐시 생성 토큰",
+	"entity.RequestLogStep.cache_creation_1h_tokens": "1시간 TTL 캐시 생성 토큰",
+	"entity.RequestLogStep.cost_usd": "step 비용 (micro-USD)",
+	"entity.RequestLogStep.cost_source": "비용 출처(provider/pricing_table)",
 	"entity.RequestLogStep.duration_ms": "소요시간(ms)",
 	"entity.RequestLogStep.ttft_ms": "Gen TTFT(ms)",
 	"entity.RequestLogStep.finish_reason": "완료 사유",

package/bundle/web-dist/server/assets/{services.generated-DyZUzXiq.js → services.generated-C8q1M41a.js}

@@ -1,4 +1,4 @@
-import { B as resolveStaleTime, C as infiniteQueryBehavior, D as notifyManager, G as timeoutManager, H as shouldThrowError, I as noop, K as focusManager, N as isValidTimeout, O as pendingThenable, R as replaceData, S as hasPreviousPage, T as fetchState, V as shallowEqualObjects, W as timeUntilStale, Z as __toESM, _ as require_jsx_runtime, a as fetch, b as getDefaultState, f as require_lib, g as useQueryClient, i as dedupeAndFlatten, j as hashKey, k as environmentManager, o as useRefreshable, q as Subscribable, s as useSSEStream, v as require_react, x as hasNextPage, z as resolveEnabled } from "./sd.generated-CE6D5clm.js";
+import { B as resolveStaleTime, C as infiniteQueryBehavior, D as notifyManager, G as timeoutManager, H as shouldThrowError, I as noop, K as focusManager, N as isValidTimeout, O as pendingThenable, R as replaceData, S as hasPreviousPage, T as fetchState, V as shallowEqualObjects, W as timeUntilStale, Z as __toESM, _ as require_jsx_runtime, a as fetch, b as getDefaultState, f as require_lib, g as useQueryClient, i as dedupeAndFlatten, j as hashKey, k as environmentManager, o as useRefreshable, q as Subscribable, s as useSSEStream, v as require_react, x as hasNextPage, z as resolveEnabled } from "./sd.generated-qzuRe35X.js";
 //#region ../../node_modules/.pnpm/@tanstack+query-core@5.95.2/node_modules/@tanstack/query-core/build/modern/queryObserver.js
 var QueryObserver = class extends Subscribable {
 	constructor(client, options) {

package/bundle/web-dist/server/assets/{show-BDul8xLb.js → show-B3q8_GDh.js}

@@ -1,6 +1,6 @@
-import { J as __commonJSMin, Y as __exportAll, Z as __toESM, _ as require_jsx_runtime, v as require_react } from "./sd.generated-CE6D5clm.js";
-import { i as RequestLogStepService, r as RequestLogService } from "./services.generated-DyZUzXiq.js";
-import { a as Link } from "./lazyRouteComponent-BlfOh9S3.js";
+import { J as __commonJSMin, Y as __exportAll, Z as __toESM, _ as require_jsx_runtime, v as require_react } from "./sd.generated-qzuRe35X.js";
+import { i as RequestLogStepService, r as RequestLogService } from "./services.generated-C8q1M41a.js";
+import { a as Link } from "./lazyRouteComponent-Cu6IkL8A.js";
 import { t as Route } from "../entry-server.generated.js";
 import { n as formatMicroUsd, t as cacheHitRate } from "./cost-CSH_t17I.js";
 import minpath from "node:path";
@@ -25308,7 +25308,11 @@ function HeaderBar({ data }) {
 		children: [
 			/* @__PURE__ */ (0, import_jsx_runtime.jsx)("span", {
 				className: "text-[15px] font-semibold text-sand-900",
-				children: data.model_name ?? "Unknown model"
+				children: data.requested_model_name !== null && data.model_name !== null && data.requested_model_name !== data.model_name ? `${data.requested_model_name} → ${data.model_name}` : data.model_name ?? "Unknown model"
+			}),
+			(data.fallback_count ?? 0) > 0 && /* @__PURE__ */ (0, import_jsx_runtime.jsxs)("span", {
+				className: "text-[10px] px-1.5 py-0.5 rounded-full bg-caution-400/15 text-caution-500 font-medium uppercase",
+				children: ["fallback ×", data.fallback_count]
 			}),
 			status && status !== "succeeded" && /* @__PURE__ */ (0, import_jsx_runtime.jsx)("span", {
 				className: `text-[10px] px-1.5 py-0.5 rounded-full font-medium uppercase ${STATUS_STYLE[status] ?? "bg-sand-100 text-sand-500"}`,
@@ -25352,6 +25356,10 @@ function MetricsPanel({ data, toolCallCount }) {
 					label: "Driver Cost",
 					value: data.cost_usd !== null ? formatMicroUsd(data.cost_usd) : "—"
 				}),
+				/* @__PURE__ */ (0, import_jsx_runtime.jsx)(Metric, {
+					label: "Cost Source",
+					value: data.cost_source ?? "legacy"
+				}),
 				/* @__PURE__ */ (0, import_jsx_runtime.jsx)(Metric, {
 					label: "Image Cost",
 					value: data.image_cost_usd !== null ? formatMicroUsd(data.image_cost_usd) : "—"
@@ -25380,6 +25388,14 @@ function MetricsPanel({ data, toolCallCount }) {
 					label: "Cache Write",
 					value: formatNum(data.cache_creation_tokens)
 				}),
+				/* @__PURE__ */ (0, import_jsx_runtime.jsx)(Metric, {
+					label: "Cache Write (5m)",
+					value: data.cache_creation_5m_tokens === null ? "—" : formatNum(data.cache_creation_5m_tokens)
+				}),
+				/* @__PURE__ */ (0, import_jsx_runtime.jsx)(Metric, {
+					label: "Cache Write (1h)",
+					value: data.cache_creation_1h_tokens === null ? "—" : formatNum(data.cache_creation_1h_tokens)
+				}),
 				/* @__PURE__ */ (0, import_jsx_runtime.jsx)(Metric, {
 					label: "Cache Hit",
 					value: cacheHitRate(data)
@@ -25614,6 +25630,14 @@ function StepTreeItem({ entry, tokensPerSecEnabled }) {
 					className: "text-[10px] px-1.5 py-0.5 rounded bg-sand-100 text-sand-500 font-mono shrink-0",
 					children: generate.finishReason
 				}),
+				generate && (generate.fallbackCount ?? 0) > 0 && /* @__PURE__ */ (0, import_jsx_runtime.jsxs)("span", {
+					className: "text-[10px] px-1.5 py-0.5 rounded-full bg-caution-400/15 text-caution-500 font-medium uppercase shrink-0",
+					children: [
+						generate.requestedModelName,
+						" → ",
+						generate.modelName
+					]
+				}),
 				hasReasoning && /* @__PURE__ */ (0, import_jsx_runtime.jsx)("span", {
 					className: "text-[10px] px-1.5 py-0.5 rounded bg-blue-100 text-blue-600 font-medium shrink-0",
 					children: "reasoning"
@@ -25662,6 +25686,18 @@ function StepTreeItem({ entry, tokensPerSecEnabled }) {
 							label: "Cache Write",
 							value: formatNum(generate.cacheCreationTokens ?? 0)
 						}),
+						/* @__PURE__ */ (0, import_jsx_runtime.jsx)(CompactMetric, {
+							label: "Cache 5m / 1h",
+							value: `${generate.cacheCreation5mTokens === null ? "—" : formatNum(generate.cacheCreation5mTokens)} / ${generate.cacheCreation1hTokens === null ? "—" : formatNum(generate.cacheCreation1hTokens)}`
+						}),
+						/* @__PURE__ */ (0, import_jsx_runtime.jsx)(CompactMetric, {
+							label: "Cost",
+							value: generate.costUsd === null ? "—" : formatMicroUsd(generate.costUsd)
+						}),
+						/* @__PURE__ */ (0, import_jsx_runtime.jsx)(CompactMetric, {
+							label: "Cost Source",
+							value: generate.costSource ?? "legacy"
+						}),
 						/* @__PURE__ */ (0, import_jsx_runtime.jsx)(CompactMetric, {
 							label: "Reasoning",
 							value: generate.reasoningTokens !== null && generate.reasoningTokens !== void 0 ? formatNum(generate.reasoningTokens) : "—"
@@ -25708,10 +25744,17 @@ function buildStepTree(steps) {
 			group.generate = {
 				id: step.id,
 				stepIndex: step.step_index,
+				modelName: step.model_name,
+				requestedModelName: step.requested_model_name,
+				fallbackCount: step.fallback_count,
 				inputTokens: step.input_tokens,
 				outputTokens: step.output_tokens,
 				cacheReadTokens: step.cache_read_tokens,
 				cacheCreationTokens: step.cache_creation_tokens,
+				cacheCreation5mTokens: step.cache_creation_5m_tokens,
+				cacheCreation1hTokens: step.cache_creation_1h_tokens,
+				costUsd: step.cost_usd,
+				costSource: step.cost_source,
 				durationMs: step.duration_ms,
 				ttftMs: step.ttft_ms,
 				finishReason: step.finish_reason,

package/bundle/web-dist/server/assets/{tokens-fzMW4G74.js → tokens-Bvjn15h9.js}

@@ -1,7 +1,7 @@
-import { Z as __toESM, _ as require_jsx_runtime, g as useQueryClient, v as require_react } from "./sd.generated-CE6D5clm.js";
-import { a as TokenService, n as QgridService } from "./services.generated-DyZUzXiq.js";
+import { Z as __toESM, _ as require_jsx_runtime, g as useQueryClient, v as require_react } from "./sd.generated-qzuRe35X.js";
+import { a as TokenService, n as QgridService } from "./services.generated-C8q1M41a.js";
 import { i as Q, o as lucideKeyRound } from "../entry-server.generated.js";
-import { t as useUpdateTokenMutation } from "./use-update-token-mutation-BejUW5sw.js";
+import { t as useUpdateTokenMutation } from "./use-update-token-mutation-BlZo6swF.js";
 //#region ~icons/lucide/plus.jsx
 var import_react = /* @__PURE__ */ __toESM(require_react());
 var import_jsx_runtime = require_jsx_runtime();

package/bundle/web-dist/server/assets/{use-update-token-mutation-BejUW5sw.js → use-update-token-mutation-BlZo6swF.js}

@@ -1,4 +1,4 @@
-import { n as QgridService, o as useMutation } from "./services.generated-DyZUzXiq.js";
+import { n as QgridService, o as useMutation } from "./services.generated-C8q1M41a.js";
 //#region src/services/token/use-update-token-mutation.ts
 function useUpdateTokenMutation() {
 	return useMutation({ mutationFn: ({ id, name, quotaThreshold, weight }) => QgridService.updateToken(id, name, quotaThreshold, weight) });

package/bundle/web-dist/server/entry-server.generated.js

@@ -1,5 +1,5 @@
-import { A as functionalUpdate, B as resolveStaleTime, C as infiniteQueryBehavior, D as notifyManager, E as onlineManager, F as matchQuery, I as noop$1, J as __commonJSMin, K as focusManager, L as partialMatchKey, M as hashQueryKeyByOptions, P as matchMutation, U as skipToken, X as __require, Z as __toESM, _ as require_jsx_runtime, c as _enum, d as string, h as QueryClientProvider, j as hashKey, l as number, m as require_shim, n as SUPPORTED_LOCALES, p as require_react_dom, q as Subscribable, r as setLocale, t as SD, u as object, v as require_react, w as Query, y as Mutation } from "./assets/sd.generated-CE6D5clm.js";
-import { A as decodePath, C as findRouteMatch, D as createLRUCache, E as processRouteTree, F as last, I as replaceEqualDeep, M as findLast, N as functionalUpdate$1, O as invariant, P as isPromise, S as findFlatMatch, T as processRouteMasks, _ as interpolatePath, a as Link$1, b as trimPath, c as usePrevious$1, d as useRouter, f as getRouterContext, g as cleanPath, h as isNotFound, i as createRootRouteWithContext, j as deepEqual$2, k as createControlledPromise, l as matchContext, m as rootRouteId, n as createFileRoute, p as warning, r as createLazyFileRoute, s as useLayoutEffect$3, t as lazyRouteComponent, u as useRouterState, v as joinPaths, w as findSingleMatch, x as trimPathRight, y as resolvePath } from "./assets/lazyRouteComponent-BlfOh9S3.js";
+import { A as functionalUpdate, B as resolveStaleTime, C as infiniteQueryBehavior, D as notifyManager, E as onlineManager, F as matchQuery, I as noop$1, J as __commonJSMin, K as focusManager, L as partialMatchKey, M as hashQueryKeyByOptions, P as matchMutation, U as skipToken, X as __require, Z as __toESM, _ as require_jsx_runtime, c as _enum, d as string, h as QueryClientProvider, j as hashKey, l as number, m as require_shim, n as SUPPORTED_LOCALES, p as require_react_dom, q as Subscribable, r as setLocale, t as SD, u as object, v as require_react, w as Query, y as Mutation } from "./assets/sd.generated-qzuRe35X.js";
+import { A as decodePath, C as findRouteMatch, D as createLRUCache, E as processRouteTree, F as last, I as replaceEqualDeep, M as findLast, N as functionalUpdate$1, O as invariant, P as isPromise, S as findFlatMatch, T as processRouteMasks, _ as interpolatePath, a as Link$1, b as trimPath, c as usePrevious$1, d as useRouter, f as getRouterContext, g as cleanPath, h as isNotFound, i as createRootRouteWithContext, j as deepEqual$2, k as createControlledPromise, l as matchContext, m as rootRouteId, n as createFileRoute, p as warning, r as createLazyFileRoute, s as useLayoutEffect$3, t as lazyRouteComponent, u as useRouterState, v as joinPaths, w as findSingleMatch, x as trimPathRight, y as resolvePath } from "./assets/lazyRouteComponent-Cu6IkL8A.js";
 //#region ../../node_modules/.pnpm/@tanstack+query-core@5.95.2/node_modules/@tanstack/query-core/build/modern/hydration.js
 function defaultTransformerFn(data) {
 	return data;
@@ -22703,7 +22703,7 @@ var sections = [{
 		}
 	]
 }];
-var qgridVersion = "2.4.6";
+var qgridVersion = "2.4.8";
 function Sidebar({ className }) {
 	const pathname = useRouterState({ select: (s) => s.location.pathname });
 	const isActive = (path) => {
@@ -44584,11 +44584,11 @@ function RootComponent() {
 }
 //#endregion
 //#region src/routes/tokens.tsx
-var $$splitComponentImporter$4 = () => import("./assets/tokens-fzMW4G74.js");
+var $$splitComponentImporter$4 = () => import("./assets/tokens-Bvjn15h9.js");
 var Route$4 = createFileRoute("/tokens")({ component: lazyRouteComponent($$splitComponentImporter$4, "component") });
 //#endregion
 //#region src/routes/monit.tsx
-var $$splitComponentImporter$3 = () => import("./assets/monit-CYyY2S2f.js");
+var $$splitComponentImporter$3 = () => import("./assets/monit-DU6PLatf.js");
 var monitSearchSchema = object({ level: _enum([
 	"all",
 	"scaling",
@@ -44601,7 +44601,7 @@ var Route$3 = createFileRoute("/monit")({
 });
 //#endregion
 //#region src/routes/logs.tsx
-var $$splitComponentImporter$2 = () => import("./assets/logs-DYly4ipQ.js");
+var $$splitComponentImporter$2 = () => import("./assets/logs-Dx2knN2T.js");
 var logsSearchSchema = object({
 	token: string().optional(),
 	project: string().optional(),
@@ -44613,11 +44613,11 @@ var Route$2 = createFileRoute("/logs")({
 });
 //#endregion
 //#region src/routes/index.tsx
-var $$splitComponentImporter$1 = () => import("./assets/routes-Dqu3osXt.js");
+var $$splitComponentImporter$1 = () => import("./assets/routes-CTco1oJZ.js");
 var Route$1 = createFileRoute("/")({ component: lazyRouteComponent($$splitComponentImporter$1, "component") });
 //#endregion
 //#region src/routes/requests/show.tsx
-var $$splitComponentImporter = () => import("./assets/show-BDul8xLb.js");
+var $$splitComponentImporter = () => import("./assets/show-B3q8_GDh.js");
 var showSearchSchema = object({ id: number() });
 var Route = createFileRoute("/requests/show")({
 	validateSearch: showSearchSchema,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@cartanova/qgrid-cli",
-  "version": "2.4.6",
+  "version": "2.4.8",
   "repository": {
     "type": "git",
     "url": "https://github.com/cartanova-ai/qgrid"
@@ -40,7 +40,7 @@
     "knex": "^3.1.0",
     "pg": "^8.16.3",
     "radashi": "^12.2.0",
-    "sonamu": "^0.9.4",
+    "sonamu": "0.9.9",
     "typescript": "5.9.3",
     "zod": "^4.3.6"
   },

package/skills/qgrid/SKILL.md

@@ -32,12 +32,16 @@ When helping a user set up local `@cartanova/qgrid-ai-sdk`, `createQgridLogger`,
 
 Prefer `QGRID_PROJECT_NAME` as the project-wide default. Use config `projectName` only for a deliberate per-call/per-provider override. Do not add `projectName` or `project_name` under `providerOptions.qgrid`; current code does not read it there.
 
+When writing AI SDK examples or setup code, import the public `QgridProviderOptions` type and apply `satisfies QgridProviderOptions` to the value under `providerOptions.qgrid`. AI SDK types the outer `providerOptions` as a generic JSON record, so it cannot infer qgrid's option names and values by itself. `QgridProviderOptions` is the inner qgrid namespace type, not the outer record type.
+
 ## Non-Negotiable Boundaries
 
 - Use `packages/ai-sdk` as the active public SDK surface.
 - Treat `packages/sdk` as deprecated. Read it only for legacy context or migration clues. Do not add new examples or features on top of it unless explicitly asked for legacy work.
 - Route provider models by prefix: `openai/*` goes to the OpenAI Codex runtime, `anthropic/*` goes to the Anthropic Claude Code runtime. Prefix-less model fallback is not implemented.
+- Treat Fable 5 safety fallback as upstream Claude Code behavior, not qgrid routing: a classifier refusal can retry on Opus 4.8. Do not add a second qgrid retry. Preserve requested Fable versus actual serving Opus, fallback history, and provider-reported cost; a fresh Claude Code process means the fallback is not sticky across qgrid requests.
 - Treat dashboard work as Sonamu API/model/generated-client/web work, not isolated frontend work.
+- Never hand-author migration files for Sonamu-managed schema. Change the entity definition, inspect `sonamu migrate status`, and create migration files with `pnpm --dir packages/api sonamu migrate generate`. Inspect the generated files before applying them. If Sonamu cannot express a required schema change, stop and ask rather than silently replacing its workflow with a custom migration.
 - Keep OpenAI Codex built-in tools, apps, plugins, skills, web search, shell, and environment instruction blocks disabled unless the user explicitly asks for agentic Codex behavior.
 - Treat OpenAI image generation as an opt-in Codex `image_generation` tool path. Inspect current code before modifying it; it is not a direct Images API call and its image cost is an estimate.
 

package/skills/qgrid/references/ai-sdk-provider-contract.md

@@ -36,6 +36,25 @@ Setup agents should treat a missing project label as an actionable configuration
 
 qgrid-specific behavior is controlled through `providerOptions.qgrid`. Explain these options precisely because they are qgrid's extension point beyond normal AI SDK usage.
 
+AI SDK declares the outer `providerOptions` as a generic JSON record, so it does not infer qgrid-specific keys or literal values. qgrid exports `QgridProviderOptions` for the inner `providerOptions.qgrid` value. Import it and put the `satisfies` check at that inner boundary:
+
+```ts
+import { generateText } from "ai";
+import { qgrid, type QgridProviderOptions } from "@cartanova/qgrid-ai-sdk";
+
+const result = await generateText({
+  model: qgrid("anthropic/claude-fable-5"),
+  prompt,
+  providerOptions: {
+    qgrid: {
+      effort: "high",
+    } satisfies QgridProviderOptions,
+  },
+});
+```
+
+Do not write `providerOptions: { ... } satisfies QgridProviderOptions`: the exported type describes the nested qgrid options object, not AI SDK's outer provider-name map. Use this typed pattern in qgrid setup guidance and examples so misspelled keys and invalid option values fail at compile time.
+
 `providerOptions.qgrid` supports:
 
 - `sessionKey`: OpenAI thread reuse key. Disabled for Anthropic models.
@@ -43,7 +62,7 @@ qgrid-specific behavior is controlled through `providerOptions.qgrid`. Explain t
 - `verbosity`: OpenAI/Codex route only.
 - `reasoningSummary`: OpenAI/Codex route only.
 - `serviceTier`: OpenAI/Codex route only.
-- `fallbackModels`: reserved for future fallback routing.
+- `fallbackModels`: reserved for future qgrid server-side fallback routing. It is not the Fable 5 safety-refusal fallback, which is owned by Claude Code upstream.
 - `imageGeneration`: OpenAI/Codex non-stream only. Enables Codex's built-in `image_generation` tool for that request.
 - `imageGenerationOptions`: optional image quality/size hints and cost-estimation basis. Current supported values are `quality: "low" | "medium" | "high"` and `size: "1024x1024" | "1024x1536" | "1536x1024"`.
 
@@ -60,7 +79,7 @@ const result = await generateText({
   providerOptions: {
     qgrid: {
       serviceTier: "fast",
-    },
+    } satisfies QgridProviderOptions,
   },
 });
 ```
@@ -112,6 +131,15 @@ qgrid response content maps to AI SDK content:
 
 When `imageGeneration` was requested and the server returns no image part, the AI SDK provider throws a version-skew/error guard instead of silently accepting text-only output.
 
+For a successful Fable 5 refusal fallback:
+
+- `response.modelId` and `providerMetadata.qgrid.model` are the actual serving model, normally `claude-opus-4-8`.
+- `providerMetadata.qgrid.requestedModel` remains `claude-fable-5`.
+- `providerMetadata.qgrid.modelFallbacks` preserves the refusal route and optional category/explanation.
+- `providerMetadata.qgrid.costSource` reports whether cost came from Claude Code or qgrid's pricing table. Prefer the provider-reported combined cost for this path.
+
+Do not expose the requested Fable model as `response.modelId` after Opus served the answer, and do not map this upstream safety behavior onto the reserved `providerOptions.qgrid.fallbackModels` option.
+
 Usage maps qgrid standard usage into AI SDK V3 usage:
 
 - `inputTokens.total = input_tokens`

package/skills/qgrid/references/anthropic-claude-code-runtime.md

@@ -12,6 +12,7 @@ Use this reference before changing Anthropic provider behavior, Claude Code spaw
 - Config isolation
 - Input adaptation
 - Output adaptation
+- Fable refusal fallback
 - Structured output
 - 1M context
 
@@ -70,7 +71,7 @@ claude -p
   --model <canonical-model-or-1m-suffix>
   --system-prompt <text>                  # small system prompt
   --system-prompt-file <path>             # large system prompt
-  --thinking disabled
+  --thinking disabled                       # omitted for Fable 5
   --effort <effort-or-low>
   --disable-slash-commands
   --session-id <uuid>
@@ -83,7 +84,7 @@ Important details:
 - `--setting-sources project` plus seeded settings isolates user configuration.
 - `--system-prompt` or `--system-prompt-file` is always supplied. Omitting it would allow Claude Code default system prompt injection.
 - Large system prompts over 64 KiB are written to a temporary file to avoid argv `E2BIG`.
-- `--thinking disabled`, `MAX_THINKING_TOKENS=0`, and adaptive thinking env suppression keep thinking off.
+- `--thinking disabled`, `MAX_THINKING_TOKENS=0`, and adaptive thinking env suppression keep thinking off for existing models. Fable 5 requires always-on adaptive thinking, so qgrid omits all three suppressors for it.
 
 ## Spawn env
 
@@ -96,13 +97,13 @@ Included env:
 - `CLAUDE_CODE_OAUTH_TOKEN`
 - `CLAUDE_CONFIG_DIR`
 - `CLAUDE_CODE_DISABLE_AUTO_MEMORY=1`
-- `CLAUDE_CODE_DISABLE_ADAPTIVE_THINKING=1`
+- `CLAUDE_CODE_DISABLE_ADAPTIVE_THINKING=1` except for Fable 5
 - `CLAUDE_CODE_DISABLE_CLAUDE_MDS=1`
 - `CLAUDE_CODE_DISABLE_TERMINAL_TITLE=1`
 - `CLAUDE_CODE_DISABLE_BUNDLED_SKILLS=1`
 - `CLAUDE_CODE_DISABLE_WORKFLOWS=1`
 - `CLAUDE_CODE_ATTRIBUTION_HEADER=0`
-- `MAX_THINKING_TOKENS=0`
+- `MAX_THINKING_TOKENS=0` except for Fable 5
 - `CLAUDE_CODE_DISABLE_1M_CONTEXT=1` when model does not support qgrid's 1M path
 - `MAX_STRUCTURED_OUTPUT_RETRIES` for structured output only
 
@@ -144,8 +145,43 @@ Anthropic native usage categories are mutually exclusive. qgrid normalizes them
 inputTokens = input_tokens + cache_creation_input_tokens + cache_read_input_tokens
 cachedInputTokens = cache_read_input_tokens
 cacheCreationInputTokens = cache_creation_input_tokens
+cacheCreationInputTokens5m = cache_creation.ephemeral_5m_input_tokens
+cacheCreationInputTokens1h = cache_creation.ephemeral_1h_input_tokens
 ```
 
+## Fable refusal fallback
+
+Fable 5 runs upstream safety classifiers. A refusal is a successful HTTP response with
+`stop_reason: "refusal"`, not an HTTP/provider error, and can happen before output or after
+partial streamed output. Claude Code handles the refusal fallback and can retry on Opus 4.8.
+This is distinct from the CLI `--fallback-model` flag, which is for overload on the default
+model. qgrid does not configure either mechanism or add another retry.
+
+The stream adapter must preserve all observable routing evidence:
+
+- `system.subtype = "model_refusal_fallback"` provides the original/fallback models and refusal details.
+- `assistant.message.model` identifies the model that produced that message.
+- `usage.iterations` contains the original `message` and `fallback_message` when available.
+- Treat `stop_reason`, rather than `stop_details` or empty content, as the authoritative refusal signal. Category and explanation are optional metadata.
+- `model_refusal_no_fallback` or a terminal `stop_reason: "refusal"` is an error, even if the outer result subtype says `success`. A successfully served fallback has a non-refusal terminal reason and a `fallback_message` iteration.
+
+`QueryOutput.model` is the actual serving model. `requestedModel` and `modelFallbacks` preserve
+the caller's requested model and routing history. Because qgrid uses a fresh Claude Code process
+for each request, any Claude Code session-sticky fallback state does not carry to the next request;
+each new Fable request may independently refuse and fall back again.
+
+Cost and usage must follow the upstream combined result. Prefer Claude Code's `total_cost_usd` and
+preserve `usage.iterations` only as attribution evidence; do not independently price every iteration
+and add them together. An upstream pre-output refusal is not billed, while work completed before a
+mid-stream refusal can be billed. Upstream fallback handling applies fallback credit where applicable,
+so a qgrid-side retry would risk duplicate work and incorrect cache-write billing.
+
+Upstream references:
+
+- [Refusals and fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback)
+- [Fallback credit](https://platform.claude.com/docs/en/build-with-claude/fallback-credit)
+- [Claude Code CLI `--fallback-model`](https://code.claude.com/docs/en/cli-reference)
+
 ## Structured output
 
 qgrid strictifies schemas before provider dispatch. Claude Code `--json-schema` is not grammar-constrained decoding; it is a `StructuredOutput` tool plus post-validation behavior. Keep schemas strict and preserve error signaling for failed structured-output attempts.
@@ -156,8 +192,10 @@ qgrid strictifies schemas before provider dispatch. Claude Code `--json-schema`
 
 Model normalization strips provider prefix and `[1m]` for canonical model/cost keys.
 
-qgrid's measured 1M support set currently includes:
+qgrid's exact 1M support set currently includes:
 
+- `claude-fable-5`
+- `claude-sonnet-5`
 - `claude-sonnet-4-6`
 - `claude-opus-4-6`
 - `claude-opus-4-8`

package/skills/qgrid/references/decision-rationale.md

@@ -29,8 +29,10 @@ Key decisions:
 - `packages/ai-sdk` is the active public SDK. `packages/sdk` is deprecated and should be read only for legacy context.
 - qgrid's AI SDK provider follows the AI SDK `LanguageModelV3` contract. Generic usage belongs to AI SDK docs; qgrid-specific docs should focus on qgrid routing, logging, cache/session behavior, and `providerOptions.qgrid`.
 - qgrid-specific options live under `providerOptions.qgrid`, not under provider-specific namespaces such as `providerOptions.openai`. Consumers should not need to branch on Codex versus Claude internals.
+- AI SDK exposes the outer `providerOptions` as a generic JSON record. `QgridProviderOptions` intentionally types the nested `providerOptions.qgrid` value, and examples should apply `satisfies` there to restore qgrid-specific inference without replacing AI SDK's outer contract.
 - `projectName` is provider config/payload camelCase. `project_name` is the Sonamu/database request-log column. Prefer `QGRID_PROJECT_NAME` as the project-wide default so request logs and metrics remain filterable at volume.
 - `fallbackModels` may exist as a typed future option, but it is not a server wire contract until the server implements fallback routing. Do not forward it as behavior just because the type exists.
+- Fable 5 refusal fallback is an upstream Claude Code safety path to Opus 4.8, not qgrid's reserved `fallbackModels` feature or Claude Code's overload-only `--fallback-model` flag. qgrid observes and reports the actual route and provider cost; it must not retry again.
 - The SDK should stay a thin adapter. Server APIs own request-log lifecycle, provider dispatch, structured-output emulation, and provider runtime details.
 
 ## Tool Calling And Request-Log Lifecycle

package/skills/qgrid/references/prompt-cache-and-usage.md

@@ -97,7 +97,10 @@ Do not double-count cache subfields. This was a real dashboard bug: OpenAI cache
 - OpenAI cost normally uses qgrid's price table.
 - OpenAI image generation stores a separate `image_cost_usd` estimate. It assumes Codex's image tool is priced as `gpt-image-2` for the selected/default quality and size, because qgrid does not receive exact image tool usage from Codex.
 - Anthropic Claude Code may return `total_cost_usd`; qgrid prefers provider cost when present and positive.
-- Anthropic cache creation uses 5-minute cache write price in fallback tables.
+- Anthropic responses can split cache creation into `ephemeral_5m_input_tokens` and `ephemeral_1h_input_tokens`; preserve that split and price each TTL separately.
+- Claude Code automatically requests 1-hour cache TTL for subscription OAuth and 5-minute TTL for API-key or usage-credit paths. Only legacy responses without a TTL breakdown use the subscription path's 1-hour rate as qgrid's fallback.
+- Persist per-step `cost_usd`, `cost_source`, and the 5m/1h split. New request logs use that exact stored cost; only legacy rows with no `cost_source` are repriced from the current table.
+- When Fable refuses and Claude Code serves Opus, prefer Claude Code's positive `total_cost_usd`. Persist Fable as the requested model and Opus as the serving model instead of pricing the combined turn as Fable.
 - OpenAI long-context surcharge applies to the whole request when input exceeds the model threshold, not just the excess tokens.
 
 When changing models or prices, verify against current official pricing before editing the table.

package/skills/qgrid/references/request-log-run-lifecycle.md

@@ -47,14 +47,18 @@ Stored request log usage uses qgrid-standard semantics:
 - `input_tokens`: total input, including cache read/write.
 - `cache_read_tokens`: cached input read.
 - `cache_creation_tokens`: prompt cache write input.
+- `cache_creation_5m_tokens` / `cache_creation_1h_tokens`: nullable Anthropic TTL breakdown.
 - `output_tokens`: output tokens.
 - `cost_usd`: integer micro-USD in DB; displayed USD is `cost_usd / 1_000_000`.
+- `cost_source`: `provider`, `pricing_table`, or `mixed`. New rows keep the cost calculated at request time.
+- `requested_model_name` / `model_name`: requested and actual serving models; they differ on Fable refusal fallback.
+- `fallback_count`: number of observed model fallbacks in the run/step.
 - `image_cost_usd`: integer micro-USD estimate for Codex image generation output.
 - `image_cost_method`: string such as `assumed:gpt-image-2:medium:1536x1024:png`.
 
 OpenAI/Codex: use per-turn `.last` usage from `thread/tokenUsage/updated`.
 
-Anthropic: normalize native mutually exclusive categories by summing input + cache creation + cache read into total input.
+Anthropic: normalize native mutually exclusive categories by summing input + cache creation + cache read into total input. Prefer positive Claude Code `total_cost_usd`; otherwise calculate from the actual serving model and TTL split.
 
 For image requests, keep `cost_usd` as the Codex driver model token cost. Image output cost is separate because qgrid observes Codex's `image_generation` result, not the OpenAI Images API usage object. Treat `image_cost_usd` as a price-table estimate that may be inaccurate if Codex changes its underlying image accounting.
 
@@ -62,7 +66,7 @@ Reference images for image generation are not stored in `request_logs.user_promp
 
 ## Legacy normalization
 
-`RequestLogModel` normalizes legacy Anthropic rows where stored `input_tokens` may not include cache read/write. Keep this in mind when changing cost or display logic.
+`RequestLogModel` normalizes legacy Anthropic rows where stored `input_tokens` may not include cache read/write. Rows with `cost_source = NULL` are legacy and can be repriced from the current table; rows with a source retain their exact stored cost across price/promotion changes.
 
 ## TTFT
 

package/skills/qgrid/references/sonamu-api-web-flow.md

@@ -21,6 +21,19 @@ qgrid dashboard work is usually a Sonamu API/model/generated-client/web change,
 - Web components under `packages/web/src/components/qgrid` consume generated API shapes.
 - The main dashboard sidebar version should stay synced to `packages/cli/package.json`. `packages/web/vite.config.ts` reads that package version and exposes it as `__QGRID_CLI_VERSION__`; `Sidebar.tsx` displays it.
 
+## Migration workflow
+
+Sonamu entity definitions are the source of truth for managed database schema.
+
+1. Change the relevant `*.entity.json` files.
+2. Run `pnpm --dir packages/api sonamu migrate status` and inspect `preparedCodes`.
+3. Run `pnpm --dir packages/api sonamu migrate generate`.
+4. Inspect the generated, table-scoped migration files without rewriting them by hand.
+5. Apply them through Sonamu's migration target flow or qgrid's startup migration runner.
+6. Re-run migration status and require `pending: []` and `preparedCodes: []`.
+
+Do not hand-author a migration that Sonamu can derive from entity changes. If generation cannot represent the required operation, stop and make the exception explicit before changing migration code or database state. Never delete or edit a migration that has already shipped to another environment.
+
 ## Request log dashboard
 
 Request log changes usually touch: