diff --git a/.archive/README.md b/.archive/README.md
index ad5e6145da..6094dd301e 100644
--- a/.archive/README.md
+++ b/.archive/README.md
@@ -135,3 +135,49 @@ Note that the repo's own `npm run check:unused-exports` does **not** find these.
 **Verification:** `tsc --noEmit` clean, full vitest suite green (701 files / 6390 tests), production webpack build clean.
 
 **To restore:** reverse the `git mv` for the file in question. No other edits are needed.
+
+---
+
+## SWE-bench "Benchmark (Beta)" UI — archived 2026-08-16
+
+The SWE-bench Pro benchmark runner UI (task browser, run builder, per-run
+session group in the chat panel, `benchmark` WorkStation tab) is parked. On
+`develop` it was already unreachable from normal UI: no menu offered the
+`benchmark` create target, nothing created a `benchmark` tab except the E2E
+seed helpers, and `BenchmarkTabSidebar` had no consumer. The only live entry
+was clicking a legacy "Benchmark run coordinator" session in the sidebar,
+which routed to the run-list surface. **Not** the Housekeeper _token_ benchmark
+(`housekeeperTokenBenchmark`, `integrations:housekeeper.benchmark.*`) — that
+is a different feature and stays live.
+
+**What moved here (self-contained to the feature):**
+
+- `src/features/BenchmarkPanel/` — panel, task selector, `useBenchmarkTasks`, `useBenchmarkAgentBatchRun`
+- `src/modules/WorkStation/shared/SidebarModules/Benchmark/` — `BenchmarkTabSidebar`
+- `src/modules/WorkStation/TabContent/renderers/benchmark.tsx` — the `benchmark` tab renderer
+- `src/engines/ChatPanel/panels/BenchmarkRunBuilder.tsx` — the run-builder creator surface
+- `src/store/benchmark/` — batch status / active batch atoms
+- `src/api/tauri/benchmark/` — the `benchmarkApi` client over the `benchmark_*` Tauri commands
+- `src/app/root/e2e/helpers/benchmark.ts` — E2E seed/inspect helpers
+- `tests/e2e/specs/core/{benchmark-run-ui,benchmark-docker-execution}.spec.mjs` (mirrored under `.archive/tests/`)
+
+**What deliberately stayed live:**
+
+- `src-tauri/src/benchmark/` — the Rust runner/commands still compile and are registered; they are now unreferenced from the frontend and can be removed in a backend-only PR
+- `src/config/agentIcons.tsx` `flask-conical` entry and `src/assets/fileTypeIcons/folder-benchmark*.svg` — generic icon registry / file-icon theme, not feature-specific
+- Housekeeper token benchmark (`src/modules/MainApp/Integrations/Housekeeper/HousekeeperCategoryView.tsx`, `rpc.validation.housekeeperTokenBenchmark`)
+
+**Shared files edited in place** to sever the branch:
+
+- `src/store/workstation/tabs/{types.ts,tabFactory.ts,storage.ts,index.ts,factories/{index,codeEditor}.ts}` — dropped `"benchmark"` from `WorkStationTabType`, its host mapping and persisted-type allow-list, and `BenchmarkTabData` / `benchmarkTabFactory` / `createBenchmarkTab`
+- `src/modules/WorkStation/TabContent/registry.ts`, `shared/SidebarModules/index.ts`, `shared/TabBar/components/SortableTab/index.tsx` (`BookLock` icon branch), `AppShell/CodeSidebarHeaderActions.tsx`
+- `src/store/ui/chatPanelAtom.ts`, `src/types/ui/chatPanel.ts`, `src/engines/ChatPanel/navigation/chatPanelSurfaceReducer.ts` — removed `CHAT_PANEL_CREATE_TARGET.BENCHMARK`, `CHAT_PANEL_CONTENT_MODE.BENCHMARK_SESSION_GROUP`, `CHAT_PANEL_SURFACE_KIND.BENCHMARK_SESSION_GROUP` and their navigate/reducer cases
+- `src/engines/ChatPanel/{index.tsx,ChatPanelContent.tsx,ChatPanelEmptyContent.tsx,hooks/useChatPanelContentState.tsx}` — removed the run-list mount, the run-builder creator branch, and `showBenchmarkSessionGroupContent`
+- `src/scaffold/NavigationSidebar/connectors/{useWorkstationSidebarHandlers.ts,useSessionMenuItems/{index.tsx,menuItemBuilders.tsx},WorkstationSidebarConnector/sidebarConnector.pinnedAndRevealData.ts}` — removed coordinator-session routing, child-session hiding, and master-row highlighting
+- `src/util/session/sessionDisplayMetadata.ts` — removed the `benchmark` flag / `flask-conical` icon override
+- `src/app/root/{E2EBootstrap.tsx,e2e/types.ts}`, `tests/e2e/wdio.conf.mjs` — removed helper wiring and the docker fixture builder
+- `src/i18n/locales/*/sessions.json` — removed `creator.benchmark.*` and `creator.createTarget.benchmark` (13 locales)
+
+**Behavior change:** legacy "Benchmark run coordinator" sessions and their child sessions now appear in the sidebar as ordinary sessions (previously the children were hidden and the coordinator opened the run list). Any persisted `benchmark` WorkStation tab is dropped by the storage allow-list on load.
+
+**To restore:** reverse the `git mv`s above and revert the in-place edits (see the archival commit).
diff --git a/src/api/tauri/benchmark/index.ts b/.archive/src/api/tauri/benchmark/index.ts
similarity index 100%
rename from src/api/tauri/benchmark/index.ts
rename to .archive/src/api/tauri/benchmark/index.ts
diff --git a/src/app/root/e2e/helpers/benchmark.ts b/.archive/src/app/root/e2e/helpers/benchmark.ts
similarity index 100%
rename from src/app/root/e2e/helpers/benchmark.ts
rename to .archive/src/app/root/e2e/helpers/benchmark.ts
diff --git a/src/engines/ChatPanel/panels/BenchmarkRunBuilder.tsx b/.archive/src/engines/ChatPanel/panels/BenchmarkRunBuilder.tsx
similarity index 100%
rename from src/engines/ChatPanel/panels/BenchmarkRunBuilder.tsx
rename to .archive/src/engines/ChatPanel/panels/BenchmarkRunBuilder.tsx
diff --git a/src/features/BenchmarkPanel/BenchmarkTaskSelector.tsx b/.archive/src/features/BenchmarkPanel/BenchmarkTaskSelector.tsx
similarity index 100%
rename from src/features/BenchmarkPanel/BenchmarkTaskSelector.tsx
rename to .archive/src/features/BenchmarkPanel/BenchmarkTaskSelector.tsx
diff --git a/src/features/BenchmarkPanel/hooks/useBenchmarkAgentBatchRun.ts b/.archive/src/features/BenchmarkPanel/hooks/useBenchmarkAgentBatchRun.ts
similarity index 100%
rename from src/features/BenchmarkPanel/hooks/useBenchmarkAgentBatchRun.ts
rename to .archive/src/features/BenchmarkPanel/hooks/useBenchmarkAgentBatchRun.ts
diff --git a/src/features/BenchmarkPanel/hooks/useBenchmarkTasks.ts b/.archive/src/features/BenchmarkPanel/hooks/useBenchmarkTasks.ts
similarity index 100%
rename from src/features/BenchmarkPanel/hooks/useBenchmarkTasks.ts
rename to .archive/src/features/BenchmarkPanel/hooks/useBenchmarkTasks.ts
diff --git a/src/features/BenchmarkPanel/index.tsx b/.archive/src/features/BenchmarkPanel/index.tsx
similarity index 100%
rename from src/features/BenchmarkPanel/index.tsx
rename to .archive/src/features/BenchmarkPanel/index.tsx
diff --git a/src/modules/WorkStation/TabContent/renderers/benchmark.tsx b/.archive/src/modules/WorkStation/TabContent/renderers/benchmark.tsx
similarity index 100%
rename from src/modules/WorkStation/TabContent/renderers/benchmark.tsx
rename to .archive/src/modules/WorkStation/TabContent/renderers/benchmark.tsx
diff --git a/src/modules/WorkStation/shared/SidebarModules/Benchmark/BenchmarkTabSidebar.tsx b/.archive/src/modules/WorkStation/shared/SidebarModules/Benchmark/BenchmarkTabSidebar.tsx
similarity index 100%
rename from src/modules/WorkStation/shared/SidebarModules/Benchmark/BenchmarkTabSidebar.tsx
rename to .archive/src/modules/WorkStation/shared/SidebarModules/Benchmark/BenchmarkTabSidebar.tsx
diff --git a/src/modules/WorkStation/shared/SidebarModules/Benchmark/index.ts b/.archive/src/modules/WorkStation/shared/SidebarModules/Benchmark/index.ts
similarity index 100%
rename from src/modules/WorkStation/shared/SidebarModules/Benchmark/index.ts
rename to .archive/src/modules/WorkStation/shared/SidebarModules/Benchmark/index.ts
diff --git a/src/store/benchmark/index.ts b/.archive/src/store/benchmark/index.ts
similarity index 100%
rename from src/store/benchmark/index.ts
rename to .archive/src/store/benchmark/index.ts
diff --git a/tests/e2e/specs/core/benchmark-docker-execution.spec.mjs b/.archive/tests/e2e/specs/core/benchmark-docker-execution.spec.mjs
similarity index 100%
rename from tests/e2e/specs/core/benchmark-docker-execution.spec.mjs
rename to .archive/tests/e2e/specs/core/benchmark-docker-execution.spec.mjs
diff --git a/tests/e2e/specs/core/benchmark-run-ui.spec.mjs b/.archive/tests/e2e/specs/core/benchmark-run-ui.spec.mjs
similarity index 100%
rename from tests/e2e/specs/core/benchmark-run-ui.spec.mjs
rename to .archive/tests/e2e/specs/core/benchmark-run-ui.spec.mjs
diff --git a/src/app/root/E2EBootstrap.tsx b/src/app/root/E2EBootstrap.tsx
index 778e21054b..f279556b7e 100644
--- a/src/app/root/E2EBootstrap.tsx
+++ b/src/app/root/E2EBootstrap.tsx
@@ -61,7 +61,6 @@ import {
   removeAccount,
 } from "./e2e/helpers/accounts";
 import { createAgentOrgHelpers } from "./e2e/helpers/agentOrgs";
-import { createBenchmarkE2EHelpers } from "./e2e/helpers/benchmark";
 import { createCloudHelpers } from "./e2e/helpers/cloud";
 import { createConfigHelpers } from "./e2e/helpers/config";
 import { createDebugEndpointHelpers } from "./e2e/helpers/debugEndpoints";
@@ -305,13 +304,6 @@ export const E2EBootstrap: FC = () => {
       openAgentStationDiff,
     } = createNavigationHelpers(store);
 
-    const {
-      seedBenchmarkRun,
-      inspectBenchmarkRun,
-      startLocalDockerBenchmarkRun,
-      getBenchmarkRunStatus,
-    } = createBenchmarkE2EHelpers(store);
-
     const {
       cloudSeedAuthState,
       cloudClearAuthState,
@@ -512,10 +504,6 @@ export const E2EBootstrap: FC = () => {
       openAgentTab,
       openOrgTab,
       inspectWorkstationSurface,
-      seedBenchmarkRun,
-      inspectBenchmarkRun,
-      startLocalDockerBenchmarkRun,
-      getBenchmarkRunStatus,
       cloudSeedAuthState,
       cloudClearAuthState,
       cloudReadAuthState,
diff --git a/src/app/root/e2e/types.ts b/src/app/root/e2e/types.ts
index f13851c6b9..cca9c57a32 100644
--- a/src/app/root/e2e/types.ts
+++ b/src/app/root/e2e/types.ts
@@ -820,25 +820,6 @@ export interface E2EHelpers {
       agentConfigRootCount: number;
     }>
   >;
-  seedBenchmarkRun: (opts: {
-    batchId?: string;
-    sourcePath: string;
-    taskIds: string[];
-    activeTaskId?: string;
-  }) => Promise<Result<{ batchId: string; activeTaskId: string | null }>>;
-  inspectBenchmarkRun: () => Promise<
-    Result<{
-      batchStatus: Json | null;
-      activeBatchId: string | null;
-      activeTaskId: string | null;
-    }>
-  >;
-  startLocalDockerBenchmarkRun: (opts: {
-    sourcePath: string;
-    taskId: string;
-    patch: string;
-  }) => Promise<Result<{ status: Json }>>;
-  getBenchmarkRunStatus: (runId: string) => Promise<Result<{ status: Json }>>;
   seedUserPresence: (opts: {
     roles: CustomRoleDefinition[];
     presence: UserPresenceState;
diff --git a/src/engines/ChatPanel/ChatPanelContent.test.ts b/src/engines/ChatPanel/ChatPanelContent.test.ts
index ce82774240..e2ec681f0a 100644
--- a/src/engines/ChatPanel/ChatPanelContent.test.ts
+++ b/src/engines/ChatPanel/ChatPanelContent.test.ts
@@ -9,10 +9,6 @@ vi.mock("./SessionContentView", () => ({
     createElement("div", { "data-gui-session": sessionId }),
 }));
 
-vi.mock("@src/features/BenchmarkPanel", () => ({
-  BenchmarkPanel: () => createElement("div", { "data-benchmark": "true" }),
-}));
-
 const { ChatPanelContent } = await import("./ChatPanelContent");
 
 function render(sessionViewMode: SessionViewMode): string {
@@ -25,7 +21,6 @@ function render(sessionViewMode: SessionViewMode): string {
       onSessionContinuation: () => undefined,
       paginationEnabled: false,
       position: "right" as const,
-      showBenchmarkSessionGroupContent: false,
       showPanelContent: true,
       showSessionContent: true,
       sessionViewMode,
diff --git a/src/engines/ChatPanel/ChatPanelContent.tsx b/src/engines/ChatPanel/ChatPanelContent.tsx
index 691a72cdef..c4541bccdd 100644
--- a/src/engines/ChatPanel/ChatPanelContent.tsx
+++ b/src/engines/ChatPanel/ChatPanelContent.tsx
@@ -1,4 +1,4 @@
-import React, { Suspense } from "react";
+import React from "react";
 
 import type { SessionContinuation } from "@src/store/session/sessionTabPlacementAtom";
 import type { ChatHistoryDisplayMode } from "@src/store/ui/chatPanelAtom";
@@ -6,12 +6,6 @@ import type { ChatHistoryDisplayMode } from "@src/store/ui/chatPanelAtom";
 import SessionContentView from "./SessionContentView";
 import type { SessionViewMode } from "./hooks/useSessionViewMode";
 
-const BenchmarkPanel = React.lazy(() =>
-  import("@src/features/BenchmarkPanel").then((module) => ({
-    default: module.BenchmarkPanel,
-  }))
-);
-
 interface ChatPanelContentProps {
   currentSessionId: string | null;
   emptyChatContent: React.ReactNode;
@@ -20,7 +14,6 @@ interface ChatPanelContentProps {
   displayMode: ChatHistoryDisplayMode;
   paginationEnabled: boolean;
   position: "left" | "right";
-  showBenchmarkSessionGroupContent: boolean;
   showPanelContent: boolean;
   showSessionContent: boolean;
   /** Non-GUI surface for the active session; mounted only while one is on. */
@@ -32,8 +25,8 @@ interface ChatPanelContentProps {
 }
 
 /**
- * The shared "chat column": session transcript, the benchmark run-list (still
- * contentMode-driven), and the Launchpad / creator surfaces (`emptyChatContent`).
+ * The shared "chat column": session transcript and the Launchpad / creator
+ * surfaces (`emptyChatContent`).
  * The workspace / organization / work-item / project / explore
  * surfaces are no longer rendered here — they are dedicated tab-typed renderers
  * dispatched by `UnifiedChatPanelTabContent`.
@@ -46,7 +39,6 @@ export function ChatPanelContent({
   displayMode,
   paginationEnabled,
   position,
-  showBenchmarkSessionGroupContent,
   showPanelContent,
   showSessionContent,
   alternateSessionView,
@@ -56,11 +48,7 @@ export function ChatPanelContent({
   const alternateActive = sessionViewMode !== "gui";
   return (
     <div className="relative flex min-h-0 min-w-0 flex-1 flex-col overflow-hidden">
-      {!showPanelContent ? null : showBenchmarkSessionGroupContent ? (
-        <Suspense fallback={null}>
-          <BenchmarkPanel surface="runList" />
-        </Suspense>
-      ) : showSessionContent && currentSessionId ? (
+      {!showPanelContent ? null : showSessionContent && currentSessionId ? (
         <>
           {/* Kept mounted while another view is showing: unmounting would drop
               the virtualized chat list's measurement cache and force a full
diff --git a/src/engines/ChatPanel/ChatPanelEmptyContent.tsx b/src/engines/ChatPanel/ChatPanelEmptyContent.tsx
index 016fc3aced..b8600dd56a 100644
--- a/src/engines/ChatPanel/ChatPanelEmptyContent.tsx
+++ b/src/engines/ChatPanel/ChatPanelEmptyContent.tsx
@@ -41,11 +41,6 @@ const CreateWorkItemView = React.lazy(
   () =>
     import("@src/modules/ProjectManager/WorkItems/components/CreateWorkItemView")
 );
-const BenchmarkRunBuilder = React.lazy(() =>
-  import("./panels/BenchmarkRunBuilder").then((module) => ({
-    default: module.BenchmarkRunBuilder,
-  }))
-);
 
 type SessionCreatorSlot = NonNullable<ChatPanelProps["sessionCreatorSlot"]>;
 type SessionCreatorSlotProps = React.ComponentProps<SessionCreatorSlot>;
@@ -421,13 +416,5 @@ export function ChatPanelEmptyContent({
     return renderCollabOrgCreator();
   }
 
-  if (createTarget === CHAT_PANEL_CREATE_TARGET.BENCHMARK) {
-    return (
-      <Suspense fallback={null}>
-        <BenchmarkRunBuilder className={creatorClassName} />
-      </Suspense>
-    );
-  }
-
   return null;
 }
diff --git a/src/engines/ChatPanel/TabContent/UnifiedChatPanelTabContent.tsx b/src/engines/ChatPanel/TabContent/UnifiedChatPanelTabContent.tsx
index d5b7dde909..288485f976 100644
--- a/src/engines/ChatPanel/TabContent/UnifiedChatPanelTabContent.tsx
+++ b/src/engines/ChatPanel/TabContent/UnifiedChatPanelTabContent.tsx
@@ -12,8 +12,8 @@ const WorkManagement = React.lazy(
 
 interface UnifiedChatPanelTabContentProps {
   activeTab: ChatPanelTab | null;
-  /** The shared "chat column" node (session transcript, Launchpad / creators,
-   *  and the benchmark run-list). Built by the host so this dispatcher stays
+  /** The shared "chat column" node (session transcript, Launchpad / creators).
+   *  Built by the host so this dispatcher stays
    *  agnostic of its heavy prop surface. */
   chatColumn: React.ReactNode;
   isTerminalTabActive: boolean;
diff --git a/src/engines/ChatPanel/TabContent/registry.ts b/src/engines/ChatPanel/TabContent/registry.ts
index 35e1f4bb82..11a30d8bfa 100644
--- a/src/engines/ChatPanel/TabContent/registry.ts
+++ b/src/engines/ChatPanel/TabContent/registry.ts
@@ -7,8 +7,8 @@
  * adding a new tab type is a type error until it has an entry — the same
  * guarantee the WorkStation `REGISTRY` provides.
  *
- * Session and Launchpad share the "chat column" (transcript / creators /
- * benchmark run-list, still contentMode-driven inside that column). Kanban and
+ * Session and Launchpad share the "chat column" (transcript / creators,
+ * still contentMode-driven inside that column). Kanban and
  * terminals render in their own keep-alive layers. Every other surface renders
  * a dedicated, self-sufficient component that reads its tab payload directly.
  */
diff --git a/src/engines/ChatPanel/hooks/useChatPanelContentState.tsx b/src/engines/ChatPanel/hooks/useChatPanelContentState.tsx
index 538de0b064..db5e027483 100644
--- a/src/engines/ChatPanel/hooks/useChatPanelContentState.tsx
+++ b/src/engines/ChatPanel/hooks/useChatPanelContentState.tsx
@@ -21,7 +21,6 @@ interface UseChatPanelContentStateOptions {
 }
 
 export interface ChatPanelContentState {
-  showBenchmarkSessionGroupContent: boolean;
   showCloudOrgContent: boolean;
   showExploreContent: boolean;
   showExplicitNonSessionContent: boolean;
@@ -45,38 +44,26 @@ export function useChatPanelContentState({
   selectedWorkItem,
   selectedWorkspace,
 }: UseChatPanelContentStateOptions): ChatPanelContentState {
-  const showBenchmarkSessionGroupContent =
-    active && contentMode === CHAT_PANEL_CONTENT_MODE.BENCHMARK_SESSION_GROUP;
   const showSessionContent =
     active &&
-    !showBenchmarkSessionGroupContent &&
     contentMode === CHAT_PANEL_CONTENT_MODE.SESSION &&
     Boolean(currentSessionId);
-  const showWorkItemContent =
-    Boolean(selectedWorkItem) &&
-    !showBenchmarkSessionGroupContent &&
-    !showSessionContent;
+  const showWorkItemContent = Boolean(selectedWorkItem) && !showSessionContent;
   const showProjectContent =
-    Boolean(selectedProject) &&
-    !showBenchmarkSessionGroupContent &&
-    !showSessionContent &&
-    !showWorkItemContent;
+    Boolean(selectedProject) && !showSessionContent && !showWorkItemContent;
   const showProjectOrgContent =
     Boolean(selectedProjectOrg) &&
-    !showBenchmarkSessionGroupContent &&
     !showSessionContent &&
     !showWorkItemContent &&
     !showProjectContent;
   const showExploreContent =
     exploreOpen &&
-    !showBenchmarkSessionGroupContent &&
     !showSessionContent &&
     !showWorkItemContent &&
     !showProjectContent &&
     !showProjectOrgContent;
   const showCloudOrgContent =
     Boolean(selectedCloudOrg) &&
-    !showBenchmarkSessionGroupContent &&
     !showSessionContent &&
     !showWorkItemContent &&
     !showProjectContent &&
@@ -84,7 +71,6 @@ export function useChatPanelContentState({
     !showExploreContent;
   const showWorkspaceOverviewContent =
     Boolean(selectedWorkspace) &&
-    !showBenchmarkSessionGroupContent &&
     !showSessionContent &&
     !showWorkItemContent &&
     !showProjectContent &&
@@ -95,7 +81,6 @@ export function useChatPanelContentState({
     contentMode === CHAT_PANEL_CONTENT_MODE.NON_SESSION;
   const showPanelContent =
     active ||
-    showBenchmarkSessionGroupContent ||
     showWorkItemContent ||
     showProjectContent ||
     showProjectOrgContent ||
@@ -104,7 +89,6 @@ export function useChatPanelContentState({
     showWorkspaceOverviewContent ||
     showExplicitNonSessionContent;
   const showHeader =
-    showBenchmarkSessionGroupContent ||
     showWorkItemContent ||
     showProjectContent ||
     showProjectOrgContent ||
@@ -115,7 +99,6 @@ export function useChatPanelContentState({
     active;
 
   return {
-    showBenchmarkSessionGroupContent,
     showCloudOrgContent,
     showExploreContent,
     showExplicitNonSessionContent,
diff --git a/src/engines/ChatPanel/index.tsx b/src/engines/ChatPanel/index.tsx
index 4fab0187cf..db3a8c6e3a 100644
--- a/src/engines/ChatPanel/index.tsx
+++ b/src/engines/ChatPanel/index.tsx
@@ -660,9 +660,6 @@ const ChatPanel: React.FC<ChatPanelProps> = memo(
         onSessionContinuation={handleSessionContinuation}
         paginationEnabled={paginationEnabled}
         position={position}
-        showBenchmarkSessionGroupContent={
-          contentState.showBenchmarkSessionGroupContent
-        }
         showPanelContent={contentState.showPanelContent}
         showSessionContent={contentState.showSessionContent}
         sessionViewMode={sessionView.mode}
diff --git a/src/engines/ChatPanel/navigation/chatPanelSurfaceReducer.ts b/src/engines/ChatPanel/navigation/chatPanelSurfaceReducer.ts
index 00012b4f96..55b864160f 100644
--- a/src/engines/ChatPanel/navigation/chatPanelSurfaceReducer.ts
+++ b/src/engines/ChatPanel/navigation/chatPanelSurfaceReducer.ts
@@ -57,11 +57,6 @@ export function reduceChatPanelSurfaceCommand(
         ...next,
         contentMode: CHAT_PANEL_CONTENT_MODE.SESSION,
       };
-    case CHAT_PANEL_SURFACE_KIND.BENCHMARK_SESSION_GROUP:
-      return {
-        ...next,
-        contentMode: CHAT_PANEL_CONTENT_MODE.BENCHMARK_SESSION_GROUP,
-      };
     case CHAT_PANEL_SURFACE_KIND.NEW_PROJECT:
       return {
         ...next,
diff --git a/src/i18n/locales/de/sessions.json b/src/i18n/locales/de/sessions.json
index b5d0c099c1..5abc3fa872 100644
--- a/src/i18n/locales/de/sessions.json
+++ b/src/i18n/locales/de/sessions.json
@@ -2082,8 +2082,7 @@
       "agentSession": "Agent-Sitzung",
       "manageAgents": "Agents / Skills verwalten",
       "project": "Projekt erstellen",
-      "workItem": "Work Item erstellen",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Work Item erstellen"
     },
     "searchModels": "Modelle suchen...",
     "newItem": "Neues Element",
@@ -2262,88 +2261,6 @@
       "worktree": "Worktree",
       "lockedHint": "Worktree ist für diese Sitzung gesperrt"
     },
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Benchmark-Ordner",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "Wähle eine benchmark-Aufgabe aus und öffne dann den WorkStation-Tab, um die Details vor dem Ausführen zu prüfen.",
-      "taskSelectStubOption": "Benchmark-Aufgabe auswählen",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "Ausführungsmodus",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "Erstellt ein isoliertes Git Worktree am base commit der Aufgabe und wendet den Patch an. Offizielle Tests werden nicht ausgeführt.",
-      "localDockerDescription": "Führt den offiziellen SWE-bench Pro evaluator über lokales Docker aus.",
-      "targetRepoPathTitle": "Ziel-Repo-Pfad",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "In Worktree anwenden",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "Einzellauf",
-      "batchRun": "Batch-Ausführung",
-      "sourcePathPlaceholder": "Pfad zum Benchmark-Ordner",
-      "workingDirectory": "Arbeitsverzeichnis",
-      "workingDirectoryPlaceholder": "Pfad, in dem Agents Repos klonen oder ändern sollen",
-      "localPath": "lokaler Pfad",
-      "runType": "Ausführungstyp",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "{{selected}}/{{total}} ausgewählt",
-      "concurrency": "Parallel",
-      "taskLoadingUnsupported": "Das Laden von Aufgaben ist für diesen Benchmark noch nicht verfügbar; nutze den Preflight, um die lokale Quelle zu prüfen.",
-      "attemptTitle": "Versuch",
-      "batchProgress": "{{total}} gesamt · {{queued}} in Warteschlange · {{running}} läuft · {{launched}} gestartet · {{failed}} fehlgeschlagen · {{cancelled}} abgebrochen",
-      "sessionGroupTitle": "Benchmark-Sitzungen",
-      "sessionGroupProgress": "{{total}} gesamt · {{queued}} in Warteschlange · {{running}} läuft · {{launched}} gestartet · {{failed}} fehlgeschlagen · {{cancelled}} abgebrochen",
-      "evaluateSubmitted": "Auswerten",
-      "evaluatingSubmitted": "Patches werden ausgewertet...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session: {{sessionId}}",
-      "noSessionYet": "Noch keine Session",
-      "testResultPending": "Testergebnis ausstehend",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills & Tools",
     "agentSpotlightViewMode": "Agent-Ansichtsmodus",
     "agentSpotlightListView": "Listenansicht",
diff --git a/src/i18n/locales/en/sessions.json b/src/i18n/locales/en/sessions.json
index 193d9f3f82..583bc06cf6 100644
--- a/src/i18n/locales/en/sessions.json
+++ b/src/i18n/locales/en/sessions.json
@@ -2162,8 +2162,7 @@
       "agentSession": "Agent session",
       "manageAgents": "Manage agents / skills",
       "project": "Create project",
-      "workItem": "Create work item",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Create work item"
     },
     "searchModels": "Search models...",
     "newItem": "New Item",
@@ -2348,88 +2347,6 @@
       "cancel": "Cancel"
     },
     "viewDiff": "View diff",
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Benchmark folder",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "Choose a benchmark task, then open the WorkStation tab to preview details before running.",
-      "taskSelectStubOption": "Select a benchmark task",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "Run mode",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "Creates an isolated git Worktree at the task base commit and applies the patch. This does not run the official tests.",
-      "localDockerDescription": "Runs the official SWE-bench Pro evaluator through local Docker.",
-      "targetRepoPathTitle": "Target repo path",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "Apply in Worktree",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "Single run",
-      "batchRun": "Batch run",
-      "sourcePathPlaceholder": "Path to benchmark folder",
-      "workingDirectory": "Working directory",
-      "workingDirectoryPlaceholder": "Path where agents should clone or modify repos",
-      "localPath": "local path",
-      "runType": "Run type",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "{{selected}}/{{total}} selected",
-      "concurrency": "Parallel",
-      "taskLoadingUnsupported": "Task loading is not available for this benchmark yet; use preflight to check the local source.",
-      "attemptTitle": "Attempt",
-      "batchProgress": "{{total}} total · {{queued}} queued · {{running}} running · {{launched}} launched · {{failed}} failed · {{cancelled}} cancelled",
-      "sessionGroupTitle": "Benchmark sessions",
-      "sessionGroupProgress": "{{total}} total · {{queued}} queued · {{running}} running · {{launched}} launched · {{failed}} failed · {{cancelled}} cancelled",
-      "evaluateSubmitted": "Evaluate",
-      "evaluatingSubmitted": "Evaluating patches...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session: {{sessionId}}",
-      "noSessionYet": "No session yet",
-      "testResultPending": "Test result pending",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills & Tools",
     "agentSpotlightViewMode": "Agent view mode",
     "agentSpotlightListView": "List view",
diff --git a/src/i18n/locales/es/sessions.json b/src/i18n/locales/es/sessions.json
index a7d439f346..ef722a8a24 100644
--- a/src/i18n/locales/es/sessions.json
+++ b/src/i18n/locales/es/sessions.json
@@ -2084,8 +2084,7 @@
       "agentSession": "Sesión de Agent",
       "manageAgents": "Gestionar Agents / Skills",
       "project": "Crear proyecto",
-      "workItem": "Crear Work Item",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Crear Work Item"
     },
     "searchModels": "Buscar modelos...",
     "newItem": "Nuevo elemento",
@@ -2264,88 +2263,6 @@
       "worktree": "Worktree",
       "lockedHint": "El Worktree está bloqueado para esta sesión"
     },
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Carpeta de benchmark",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "Elige una tarea benchmark y abre la pestaña WorkStation para revisar los detalles antes de ejecutarla.",
-      "taskSelectStubOption": "Seleccionar tarea benchmark",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "Modo de ejecución",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "Crea un Git Worktree aislado en el base commit de la tarea y aplica el patch. No ejecuta las pruebas oficiales.",
-      "localDockerDescription": "Ejecuta el evaluator oficial de SWE-bench Pro con Docker local.",
-      "targetRepoPathTitle": "Ruta del Repo objetivo",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "Aplicar en Worktree",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "Ejecución única",
-      "batchRun": "Ejecución por lotes",
-      "sourcePathPlaceholder": "Ruta a la carpeta de benchmark",
-      "workingDirectory": "Directorio de trabajo",
-      "workingDirectoryPlaceholder": "Ruta donde los agentes deben clonar o modificar repositorios",
-      "localPath": "ruta local",
-      "runType": "Tipo de ejecución",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "{{selected}}/{{total}} seleccionadas",
-      "concurrency": "Paralelo",
-      "taskLoadingUnsupported": "La carga de tareas aún no está disponible para este benchmark; usa preflight para comprobar la fuente local.",
-      "attemptTitle": "Intento",
-      "batchProgress": "{{total}} en total · {{queued}} en cola · {{running}} en ejecución · {{launched}} iniciadas · {{failed}} fallidas · {{cancelled}} canceladas",
-      "sessionGroupTitle": "Sesiones de benchmark",
-      "sessionGroupProgress": "{{total}} en total · {{queued}} en cola · {{running}} en ejecución · {{launched}} iniciadas · {{failed}} fallidas · {{cancelled}} canceladas",
-      "evaluateSubmitted": "Evaluar",
-      "evaluatingSubmitted": "Evaluando parches...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session: {{sessionId}}",
-      "noSessionYet": "Aún no hay Session",
-      "testResultPending": "Resultado de prueba pendiente",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills y herramientas",
     "agentSpotlightViewMode": "Modo de vista de Agent",
     "agentSpotlightListView": "Vista de lista",
diff --git a/src/i18n/locales/fr/sessions.json b/src/i18n/locales/fr/sessions.json
index f15b7750f3..8af21c642c 100644
--- a/src/i18n/locales/fr/sessions.json
+++ b/src/i18n/locales/fr/sessions.json
@@ -2084,8 +2084,7 @@
       "agentSession": "Session Agent",
       "manageAgents": "Gérer les Agents / Skills",
       "project": "Créer un projet",
-      "workItem": "Créer un Work Item",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Créer un Work Item"
     },
     "searchModels": "Rechercher des modèles...",
     "newItem": "Nouvel élément",
@@ -2264,88 +2263,6 @@
       "worktree": "Worktree",
       "lockedHint": "Le Worktree est verrouillé pour cette session"
     },
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Dossier de benchmark",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "Choisissez une tâche benchmark, puis ouvrez l’onglet WorkStation pour prévisualiser les détails avant l’exécution.",
-      "taskSelectStubOption": "Sélectionner une tâche benchmark",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "Mode d’exécution",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "Crée un Git Worktree isolé au base commit de la tâche et applique le patch. Les tests officiels ne sont pas exécutés.",
-      "localDockerDescription": "Exécute l’evaluator officiel SWE-bench Pro via Docker local.",
-      "targetRepoPathTitle": "Chemin du Repo cible",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "Appliquer dans Worktree",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "Exécution unique",
-      "batchRun": "Exécution groupée",
-      "sourcePathPlaceholder": "Chemin vers le dossier de benchmark",
-      "workingDirectory": "Répertoire de travail",
-      "workingDirectoryPlaceholder": "Chemin où les agents doivent cloner ou modifier les dépôts",
-      "localPath": "chemin local",
-      "runType": "Type d’exécution",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "{{selected}}/{{total}} sélectionnées",
-      "concurrency": "Parallèle",
-      "taskLoadingUnsupported": "Le chargement des tâches n’est pas encore disponible pour ce benchmark ; utilisez le preflight pour vérifier la source locale.",
-      "attemptTitle": "Tentative",
-      "batchProgress": "{{total}} au total · {{queued}} en file · {{running}} en cours · {{launched}} lancées · {{failed}} échouées · {{cancelled}} annulées",
-      "sessionGroupTitle": "Sessions de benchmark",
-      "sessionGroupProgress": "{{total}} au total · {{queued}} en file · {{running}} en cours · {{launched}} lancées · {{failed}} échouées · {{cancelled}} annulées",
-      "evaluateSubmitted": "Évaluer",
-      "evaluatingSubmitted": "Évaluation des patchs...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session : {{sessionId}}",
-      "noSessionYet": "Aucune session pour le moment",
-      "testResultPending": "Résultat de test en attente",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills et outils",
     "agentSpotlightViewMode": "Mode d’affichage des Agents",
     "agentSpotlightListView": "Vue liste",
diff --git a/src/i18n/locales/ja/sessions.json b/src/i18n/locales/ja/sessions.json
index a741871004..0f935910e2 100644
--- a/src/i18n/locales/ja/sessions.json
+++ b/src/i18n/locales/ja/sessions.json
@@ -2083,8 +2083,7 @@
       "agentSession": "Agent セッション",
       "manageAgents": "Agents / Skills を管理",
       "project": "プロジェクトを作成",
-      "workItem": "Work Item を作成",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Work Item を作成"
     },
     "searchModels": "モデルを検索...",
     "newItem": "新しいアイテム",
@@ -2263,88 +2262,6 @@
       "worktree": "Worktree",
       "lockedHint": "このセッションでは Worktree がロックされています"
     },
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Benchmark フォルダ",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "benchmark タスクを選択し、実行前に WorkStation タブで詳細をプレビューします。",
-      "taskSelectStubOption": "benchmark タスクを選択",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "実行モード",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "タスクの base commit に隔離された Git Worktree を作成して patch を適用します。公式テストは実行しません。",
-      "localDockerDescription": "ローカル Docker で公式 SWE-bench Pro evaluator を実行します。",
-      "targetRepoPathTitle": "対象 Repo パス",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "Worktree に適用",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "単一実行",
-      "batchRun": "バッチ実行",
-      "sourcePathPlaceholder": "Benchmark フォルダへのパス",
-      "workingDirectory": "作業ディレクトリ",
-      "workingDirectoryPlaceholder": "Agent が Repo を clone または変更するパス",
-      "localPath": "ローカルパス",
-      "runType": "実行タイプ",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "{{selected}}/{{total}} 件選択済み",
-      "concurrency": "並列数",
-      "taskLoadingUnsupported": "この Benchmark ではタスク読み込みはまだ利用できません。preflight でローカルソースを確認してください。",
-      "attemptTitle": "Attempt",
-      "batchProgress": "合計 {{total}} · キュー {{queued}} · 実行中 {{running}} · 起動済み {{launched}} · 失敗 {{failed}} · キャンセル {{cancelled}}",
-      "sessionGroupTitle": "Benchmark sessions",
-      "sessionGroupProgress": "合計 {{total}} · キュー {{queued}} · 実行中 {{running}} · 起動済み {{launched}} · 失敗 {{failed}} · キャンセル {{cancelled}}",
-      "evaluateSubmitted": "評価",
-      "evaluatingSubmitted": "パッチを評価中...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session: {{sessionId}}",
-      "noSessionYet": "Session はまだありません",
-      "testResultPending": "テスト結果は保留中",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills とツール",
     "agentSpotlightViewMode": "Agent 表示モード",
     "agentSpotlightListView": "リスト表示",
diff --git a/src/i18n/locales/ko/sessions.json b/src/i18n/locales/ko/sessions.json
index f4a0f8bb25..fa615433bd 100644
--- a/src/i18n/locales/ko/sessions.json
+++ b/src/i18n/locales/ko/sessions.json
@@ -2083,8 +2083,7 @@
       "agentSession": "Agent 세션",
       "manageAgents": "Agents / Skills 관리",
       "project": "프로젝트 만들기",
-      "workItem": "Work Item 만들기",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Work Item 만들기"
     },
     "searchModels": "모델 검색...",
     "newItem": "새 항목",
@@ -2263,88 +2262,6 @@
       "worktree": "Worktree",
       "lockedHint": "이 세션에서 Worktree가 잠겨 있습니다"
     },
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Benchmark 폴더",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "benchmark 작업을 선택한 다음 WorkStation 탭을 열어 실행 전에 세부 정보를 미리 봅니다.",
-      "taskSelectStubOption": "benchmark 작업 선택",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "실행 모드",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "작업의 base commit에 격리된 Git Worktree를 만들고 patch를 적용합니다. 공식 테스트는 실행하지 않습니다.",
-      "localDockerDescription": "로컬 Docker로 공식 SWE-bench Pro evaluator를 실행합니다.",
-      "targetRepoPathTitle": "대상 Repo 경로",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "Worktree에 적용",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "단일 실행",
-      "batchRun": "배치 실행",
-      "sourcePathPlaceholder": "Benchmark 폴더 경로",
-      "workingDirectory": "작업 디렉터리",
-      "workingDirectoryPlaceholder": "Agent가 Repo를 클론하거나 수정할 경로",
-      "localPath": "로컬 경로",
-      "runType": "실행 유형",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "{{selected}}/{{total}}개 선택됨",
-      "concurrency": "병렬 수",
-      "taskLoadingUnsupported": "이 Benchmark에서는 아직 작업 로드를 사용할 수 없습니다. preflight로 로컬 소스를 확인하세요.",
-      "attemptTitle": "Attempt",
-      "batchProgress": "총 {{total}}개 · 대기 {{queued}}개 · 실행 중 {{running}}개 · 시작됨 {{launched}}개 · 실패 {{failed}}개 · 취소 {{cancelled}}개",
-      "sessionGroupTitle": "Benchmark sessions",
-      "sessionGroupProgress": "총 {{total}}개 · 대기 {{queued}}개 · 실행 중 {{running}}개 · 시작됨 {{launched}}개 · 실패 {{failed}}개 · 취소 {{cancelled}}개",
-      "evaluateSubmitted": "평가",
-      "evaluatingSubmitted": "패치 평가 중...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session: {{sessionId}}",
-      "noSessionYet": "아직 Session 없음",
-      "testResultPending": "테스트 결과 대기 중",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills 및 도구",
     "agentSpotlightViewMode": "Agent view mode",
     "agentSpotlightListView": "List view",
diff --git a/src/i18n/locales/pl/sessions.json b/src/i18n/locales/pl/sessions.json
index b626308843..3323149a9c 100644
--- a/src/i18n/locales/pl/sessions.json
+++ b/src/i18n/locales/pl/sessions.json
@@ -2137,8 +2137,7 @@
       "agentSession": "Sesja Agent",
       "manageAgents": "Zarządzaj Agents / Skills",
       "project": "Utwórz projekt",
-      "workItem": "Utwórz Work Item",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Utwórz Work Item"
     },
     "searchModels": "Wyszukaj modele...",
     "newItem": "Nowy element",
@@ -2317,88 +2316,6 @@
       "cancel": "Anuluj"
     },
     "viewDiff": "Wyświetl różnice",
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Folder benchmarku",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "Wybierz zadanie benchmark, a następnie otwórz kartę WorkStation, aby podejrzeć szczegóły przed uruchomieniem.",
-      "taskSelectStubOption": "Wybierz zadanie benchmark",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "Tryb uruchomienia",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "Tworzy izolowany Git Worktree na base commit zadania i stosuje patch. Nie uruchamia oficjalnych testów.",
-      "localDockerDescription": "Uruchamia oficjalny evaluator SWE-bench Pro przez lokalny Docker.",
-      "targetRepoPathTitle": "Ścieżka docelowego Repo",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "Zastosuj w Worktree",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "Pojedyncze uruchomienie",
-      "batchRun": "Uruchomienie zbiorcze",
-      "sourcePathPlaceholder": "Ścieżka do folderu benchmarku",
-      "workingDirectory": "Katalog roboczy",
-      "workingDirectoryPlaceholder": "Ścieżka, gdzie agenci mają klonować lub modyfikować repozytoria",
-      "localPath": "ścieżka lokalna",
-      "runType": "Typ uruchomienia",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "Wybrano: {{selected}}/{{total}}",
-      "concurrency": "Równolegle",
-      "taskLoadingUnsupported": "Ładowanie zadań nie jest jeszcze dostępne dla tego benchmarku; użyj preflight, aby sprawdzić lokalne źródło.",
-      "attemptTitle": "Attempt",
-      "batchProgress": "Łącznie {{total}} · {{queued}} w kolejce · {{running}} uruchomione · {{launched}} wystartowało · {{failed}} niepowodzeń · {{cancelled}} anulowano",
-      "sessionGroupTitle": "Benchmark sessions",
-      "sessionGroupProgress": "Łącznie {{total}} · {{queued}} w kolejce · {{running}} uruchomione · {{launched}} wystartowało · {{failed}} niepowodzeń · {{cancelled}} anulowano",
-      "evaluateSubmitted": "Oceń",
-      "evaluatingSubmitted": "Ocenianie poprawek...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session: {{sessionId}}",
-      "noSessionYet": "Brak Session",
-      "testResultPending": "Wynik testu oczekuje",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills i narzędzia",
     "agentSpotlightViewMode": "Tryb widoku Agent",
     "agentSpotlightListView": "Widok listy",
diff --git a/src/i18n/locales/pt/sessions.json b/src/i18n/locales/pt/sessions.json
index 2d7bbe2dc5..33edd82f3f 100644
--- a/src/i18n/locales/pt/sessions.json
+++ b/src/i18n/locales/pt/sessions.json
@@ -2099,8 +2099,7 @@
       "agentSession": "Sessão de Agent",
       "manageAgents": "Gerenciar Agents / Skills",
       "project": "Criar projeto",
-      "workItem": "Criar Work Item",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Criar Work Item"
     },
     "searchModels": "Buscar modelos...",
     "newItem": "Novo item",
@@ -2279,88 +2278,6 @@
       "cancel": "Cancelar"
     },
     "viewDiff": "Ver diff",
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Pasta de benchmark",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "Escolha uma tarefa benchmark e abra a aba WorkStation para pré-visualizar os detalhes antes de executar.",
-      "taskSelectStubOption": "Selecionar tarefa benchmark",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "Modo de execução",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "Cria um Git Worktree isolado no base commit da tarefa e aplica o patch. Não executa os testes oficiais.",
-      "localDockerDescription": "Executa o evaluator oficial do SWE-bench Pro com Docker local.",
-      "targetRepoPathTitle": "Caminho do Repo alvo",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "Aplicar no Worktree",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "Execução única",
-      "batchRun": "Execução em lote",
-      "sourcePathPlaceholder": "Caminho para a pasta de benchmark",
-      "workingDirectory": "Diretório de trabalho",
-      "workingDirectoryPlaceholder": "Caminho onde os agentes devem clonar ou modificar repositórios",
-      "localPath": "caminho local",
-      "runType": "Tipo de execução",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "{{selected}}/{{total}} selecionadas",
-      "concurrency": "Paralelo",
-      "taskLoadingUnsupported": "O carregamento de tarefas ainda não está disponível para este benchmark; use o preflight para verificar a fonte local.",
-      "attemptTitle": "Tentativa",
-      "batchProgress": "{{total}} no total · {{queued}} na fila · {{running}} em execução · {{launched}} iniciadas · {{failed}} falharam · {{cancelled}} canceladas",
-      "sessionGroupTitle": "Sessões de benchmark",
-      "sessionGroupProgress": "{{total}} no total · {{queued}} na fila · {{running}} em execução · {{launched}} iniciadas · {{failed}} falharam · {{cancelled}} canceladas",
-      "evaluateSubmitted": "Avaliar",
-      "evaluatingSubmitted": "Avaliando patches...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session: {{sessionId}}",
-      "noSessionYet": "Ainda sem Session",
-      "testResultPending": "Resultado do teste pendente",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills e ferramentas",
     "agentSpotlightViewMode": "Modo de visualização de Agent",
     "agentSpotlightListView": "Visualização em lista",
diff --git a/src/i18n/locales/ru/sessions.json b/src/i18n/locales/ru/sessions.json
index 7bf0a3ba6a..9e3a41b7ce 100644
--- a/src/i18n/locales/ru/sessions.json
+++ b/src/i18n/locales/ru/sessions.json
@@ -2127,8 +2127,7 @@
       "agentSession": "Сессия Agent",
       "manageAgents": "Управлять Agents / Skills",
       "project": "Создать проект",
-      "workItem": "Создать Work Item",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Создать Work Item"
     },
     "searchModels": "Поиск моделей...",
     "newItem": "Новый элемент",
@@ -2307,88 +2306,6 @@
       "worktree": "Worktree",
       "lockedHint": "Worktree заблокирован для этой сессии"
     },
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Папка Benchmark",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "Выберите benchmark-задачу, затем откройте вкладку WorkStation, чтобы просмотреть детали перед запуском.",
-      "taskSelectStubOption": "Выбрать benchmark-задачу",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "Режим запуска",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "Создает изолированный Git Worktree на base commit задачи и применяет patch. Официальные тесты не запускаются.",
-      "localDockerDescription": "Запускает официальный evaluator SWE-bench Pro через локальный Docker.",
-      "targetRepoPathTitle": "Путь к целевому Repo",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "Применить в Worktree",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "Одиночный запуск",
-      "batchRun": "Пакетный запуск",
-      "sourcePathPlaceholder": "Путь к папке Benchmark",
-      "workingDirectory": "Рабочий каталог",
-      "workingDirectoryPlaceholder": "Путь, где агенты должны клонировать или изменять репозитории",
-      "localPath": "локальный путь",
-      "runType": "Тип запуска",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "Выбрано: {{selected}}/{{total}}",
-      "concurrency": "Параллельно",
-      "taskLoadingUnsupported": "Загрузка задач для этого Benchmark пока недоступна; используйте preflight для проверки локального источника.",
-      "attemptTitle": "Attempt",
-      "batchProgress": "Всего {{total}} · {{queued}} в очереди · {{running}} выполняется · {{launched}} запущено · {{failed}} ошибок · {{cancelled}} отменено",
-      "sessionGroupTitle": "Benchmark sessions",
-      "sessionGroupProgress": "Всего {{total}} · {{queued}} в очереди · {{running}} выполняется · {{launched}} запущено · {{failed}} ошибок · {{cancelled}} отменено",
-      "evaluateSubmitted": "Оценить",
-      "evaluatingSubmitted": "Оценка патчей...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session: {{sessionId}}",
-      "noSessionYet": "Session пока нет",
-      "testResultPending": "Результат теста ожидается",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills и инструменты",
     "agentSpotlightViewMode": "Режим просмотра Agent",
     "agentSpotlightListView": "Список",
diff --git a/src/i18n/locales/tr/sessions.json b/src/i18n/locales/tr/sessions.json
index 8f27addab7..bacef72d6b 100644
--- a/src/i18n/locales/tr/sessions.json
+++ b/src/i18n/locales/tr/sessions.json
@@ -2084,8 +2084,7 @@
       "agentSession": "Agent oturumu",
       "manageAgents": "Agents / Skills yönet",
       "project": "Proje oluştur",
-      "workItem": "Work Item oluştur",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Work Item oluştur"
     },
     "searchModels": "Model ara...",
     "newItem": "Yeni öğe",
@@ -2264,88 +2263,6 @@
       "worktree": "Worktree",
       "lockedHint": "Bu oturum için Worktree kilitli"
     },
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Benchmark klasörü",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "Bir benchmark görevi seçin, ardından çalıştırmadan önce ayrıntıları önizlemek için WorkStation sekmesini açın.",
-      "taskSelectStubOption": "Benchmark görevi seç",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "Çalıştırma modu",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "Görevin base commit noktasında izole bir Git Worktree oluşturur ve patch uygular. Resmi testleri çalıştırmaz.",
-      "localDockerDescription": "Resmi SWE-bench Pro evaluator’ı yerel Docker ile çalıştırır.",
-      "targetRepoPathTitle": "Hedef Repo yolu",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "Worktree içinde uygula",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "Tek çalıştırma",
-      "batchRun": "Toplu çalıştırma",
-      "sourcePathPlaceholder": "Benchmark klasörünün yolu",
-      "workingDirectory": "Çalışma dizini",
-      "workingDirectoryPlaceholder": "Agent'ların repoları klonlayacağı veya değiştireceği yol",
-      "localPath": "yerel yol",
-      "runType": "Çalıştırma türü",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "{{selected}}/{{total}} seçildi",
-      "concurrency": "Paralel",
-      "taskLoadingUnsupported": "Bu benchmark için görev yükleme henüz kullanılamıyor; yerel kaynağı kontrol etmek için preflight kullanın.",
-      "attemptTitle": "Attempt",
-      "batchProgress": "Toplam {{total}} · {{queued}} kuyrukta · {{running}} çalışıyor · {{launched}} başlatıldı · {{failed}} başarısız · {{cancelled}} iptal edildi",
-      "sessionGroupTitle": "Benchmark sessions",
-      "sessionGroupProgress": "Toplam {{total}} · {{queued}} kuyrukta · {{running}} çalışıyor · {{launched}} başlatıldı · {{failed}} başarısız · {{cancelled}} iptal edildi",
-      "evaluateSubmitted": "Değerlendir",
-      "evaluatingSubmitted": "Yamalar değerlendiriliyor...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session: {{sessionId}}",
-      "noSessionYet": "Henüz Session yok",
-      "testResultPending": "Test sonucu bekleniyor",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills ve araçlar",
     "agentSpotlightViewMode": "Agent görünüm modu",
     "agentSpotlightListView": "Liste görünümü",
diff --git a/src/i18n/locales/vi/sessions.json b/src/i18n/locales/vi/sessions.json
index b3d9f80d20..c229eb3ced 100644
--- a/src/i18n/locales/vi/sessions.json
+++ b/src/i18n/locales/vi/sessions.json
@@ -2081,8 +2081,7 @@
       "agentSession": "Phiên Agent",
       "manageAgents": "Quản lý Agents / Skills",
       "project": "Tạo dự án",
-      "workItem": "Tạo Work Item",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "Tạo Work Item"
     },
     "searchModels": "Tìm kiếm mô hình...",
     "newItem": "Mục mới",
@@ -2261,88 +2260,6 @@
       "worktree": "Worktree",
       "lockedHint": "Worktree đã được khóa cho phiên này"
     },
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Thư mục benchmark",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "Chọn một tác vụ benchmark, rồi mở tab WorkStation để xem trước chi tiết trước khi chạy.",
-      "taskSelectStubOption": "Chọn tác vụ benchmark",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "Chế độ chạy",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "Tạo Git Worktree cô lập tại base commit của tác vụ và áp dụng patch. Không chạy bộ test chính thức.",
-      "localDockerDescription": "Chạy evaluator chính thức của SWE-bench Pro qua Docker cục bộ.",
-      "targetRepoPathTitle": "Đường dẫn Repo đích",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "Áp dụng trong Worktree",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "Chạy đơn",
-      "batchRun": "Chạy hàng loạt",
-      "sourcePathPlaceholder": "Đường dẫn tới thư mục benchmark",
-      "workingDirectory": "Thư mục làm việc",
-      "workingDirectoryPlaceholder": "Đường dẫn để Agent clone hoặc sửa đổi repo",
-      "localPath": "đường dẫn cục bộ",
-      "runType": "Kiểu chạy",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "Đã chọn {{selected}}/{{total}}",
-      "concurrency": "Song song",
-      "taskLoadingUnsupported": "Chưa thể tải tác vụ cho benchmark này; hãy dùng preflight để kiểm tra nguồn cục bộ.",
-      "attemptTitle": "Attempt",
-      "batchProgress": "Tổng {{total}} · {{queued}} đang chờ · {{running}} đang chạy · {{launched}} đã khởi chạy · {{failed}} thất bại · {{cancelled}} đã hủy",
-      "sessionGroupTitle": "Benchmark sessions",
-      "sessionGroupProgress": "Tổng {{total}} · {{queued}} đang chờ · {{running}} đang chạy · {{launched}} đã khởi chạy · {{failed}} thất bại · {{cancelled}} đã hủy",
-      "evaluateSubmitted": "Đánh giá",
-      "evaluatingSubmitted": "Đang đánh giá bản vá...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session: {{sessionId}}",
-      "noSessionYet": "Chưa có Session",
-      "testResultPending": "Đang chờ kết quả kiểm thử",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills và công cụ",
     "agentSpotlightViewMode": "Chế độ xem Agent",
     "agentSpotlightListView": "Dạng danh sách",
diff --git a/src/i18n/locales/zh-Hant/sessions.json b/src/i18n/locales/zh-Hant/sessions.json
index 5a32a425e6..2f69b49a5a 100644
--- a/src/i18n/locales/zh-Hant/sessions.json
+++ b/src/i18n/locales/zh-Hant/sessions.json
@@ -2098,8 +2098,7 @@
       "agentSession": "Agent 會話",
       "manageAgents": "管理 Agent / Skills",
       "project": "建立專案",
-      "workItem": "建立 Work Item",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "建立 Work Item"
     },
     "searchModels": "搜索模型...",
     "newItem": "新建專案",
@@ -2279,88 +2278,6 @@
       "cancel": "取消"
     },
     "viewDiff": "查看 Diff",
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Benchmark 資料夾",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "選擇一個 benchmark 任務，然後開啟 WorkStation 標籤頁，在執行前預覽詳情。",
-      "taskSelectStubOption": "選擇 benchmark 任務",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "執行模式",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "在任務 base commit 建立隔離 Git Worktree 並套用 patch；不會執行官方測試。",
-      "localDockerDescription": "透過本機 Docker 執行官方 SWE-bench Pro evaluator。",
-      "targetRepoPathTitle": "目標 Repo 路徑",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "套用到 Worktree",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "單次執行",
-      "batchRun": "批次執行",
-      "sourcePathPlaceholder": "Benchmark 資料夾路徑",
-      "workingDirectory": "工作目錄",
-      "workingDirectoryPlaceholder": "Agent 複製或修改 Repo 的路徑",
-      "localPath": "本機路徑",
-      "runType": "執行類型",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "已選擇 {{selected}}/{{total}} 個",
-      "concurrency": "並行數",
-      "taskLoadingUnsupported": "此 Benchmark 暫不支援載入任務；請使用 preflight 檢查本機來源。",
-      "attemptTitle": "Attempt",
-      "batchProgress": "共 {{total}} 個 · {{queued}} 個排隊 · {{running}} 個執行中 · {{launched}} 個已啟動 · {{failed}} 個失敗 · {{cancelled}} 個已取消",
-      "sessionGroupTitle": "Benchmark sessions",
-      "sessionGroupProgress": "共 {{total}} 個 · {{queued}} 個排隊 · {{running}} 個執行中 · {{launched}} 個已啟動 · {{failed}} 個失敗 · {{cancelled}} 個已取消",
-      "evaluateSubmitted": "評估",
-      "evaluatingSubmitted": "正在評估補丁...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session：{{sessionId}}",
-      "noSessionYet": "暫無 Session",
-      "testResultPending": "測試結果待定",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills 與工具",
     "agentSpotlightViewMode": "Agent 檢視模式",
     "agentSpotlightListView": "列表檢視",
diff --git a/src/i18n/locales/zh/sessions.json b/src/i18n/locales/zh/sessions.json
index d36b60a16b..d5b1fa9986 100644
--- a/src/i18n/locales/zh/sessions.json
+++ b/src/i18n/locales/zh/sessions.json
@@ -2135,8 +2135,7 @@
       "agentSession": "Agent 会话",
       "manageAgents": "管理 Agent / Skills",
       "project": "创建项目",
-      "workItem": "创建 Work Item",
-      "benchmark": "Benchmark (Beta)"
+      "workItem": "创建 Work Item"
     },
     "searchModels": "搜索模型...",
     "newItem": "新建项目",
@@ -2321,88 +2320,6 @@
       "cancel": "取消"
     },
     "viewDiff": "查看 Diff",
-    "benchmark": {
-      "title": "Benchmark",
-      "description": "Browse SWE-bench Pro tasks, choose what to run, and check the local benchmark environment before evaluation.",
-      "phaseLabel": "Task browser",
-      "sourcePath": "Benchmark 文件夹",
-      "evaluationMode": "Evaluation mode",
-      "evaluationModes": {
-        "patchOnly": "Patch only",
-        "localDocker": "Local Docker",
-        "modal": "Modal"
-      },
-      "loadTasks": "Load tasks",
-      "runPreflight": "Run preflight",
-      "loading": "Loading...",
-      "checking": "Checking...",
-      "searchPlaceholder": "Search task ID, title, or repo",
-      "loadedTasks": "{{count}} tasks loaded",
-      "emptyTasks": "No tasks found.",
-      "words": "{{count}} words",
-      "noTaskSelected": "No task selected",
-      "taskStats": "{{words}} words · {{chars}} chars",
-      "preflightTitle": "Preflight checks",
-      "ready": "Ready",
-      "needsSetup": "Needs setup",
-      "selectTaskHint": "Select a task to preview the benchmark question.",
-      "taskSelectionTitle": "Benchmark tasks",
-      "taskSelectionDescription": "选择一个 benchmark 任务，然后打开 WorkStation 标签页，在运行前预览详情。",
-      "taskSelectStubOption": "选择 benchmark 任务",
-      "openWorkstationTab": "View Benchmark",
-      "unknownRepo": "Unknown repo",
-      "runPanelTitle": "Docker run",
-      "runPanelDescription": "Evaluate patch output against {{taskId}} with local Docker.",
-      "patchInputTitle": "Patch content",
-      "patchInputPlaceholder": "Paste a unified diff patch to evaluate with SWE-bench Pro Docker.",
-      "localDocker": "Local Docker",
-      "preflightSummary": "{{ready}}/{{total}} checks ready",
-      "createRunPlan": "Create run plan",
-      "runWithDocker": "Run with Docker",
-      "runPlanTitle": "Run plan",
-      "runStatusTitle": "Run status",
-      "noRunLogs": "No logs yet",
-      "runIdLabel": "Run",
-      "processIdLabel": "PID",
-      "resultLabel": "Result",
-      "resultPassed": "Passed",
-      "resultFailed": "Failed",
-      "noParsedResult": "No parsed result yet",
-      "evaluationModeTitle": "运行模式",
-      "patchOnlyWorktree": "Patch-only Worktree",
-      "patchOnlyWorktreeDescription": "在任务 base commit 创建隔离 Git Worktree 并应用 patch；不会运行官方测试。",
-      "localDockerDescription": "通过本地 Docker 运行官方 SWE-bench Pro evaluator。",
-      "targetRepoPathTitle": "目标 Repo 路径",
-      "targetRepoPathPlaceholder": "/Users/you/Documents/GitHub/project-repo",
-      "applyInWorktree": "应用到 Worktree",
-      "worktreePathLabel": "Worktree",
-      "singleRun": "单次运行",
-      "batchRun": "批量运行",
-      "sourcePathPlaceholder": "Benchmark 文件夹路径",
-      "workingDirectory": "工作目录",
-      "workingDirectoryPlaceholder": "Agent 克隆或修改仓库的路径",
-      "localPath": "本地路径",
-      "runType": "运行类型",
-      "kindTitle": "Benchmark",
-      "kinds": {
-        "sweBenchPro": "SWE-bench Pro",
-        "terminalBench": "Terminal-Bench"
-      },
-      "selectedTasks": "已选择 {{selected}}/{{total}} 个",
-      "concurrency": "并行数",
-      "taskLoadingUnsupported": "此 Benchmark 暂不支持加载任务；请使用 preflight 检查本地来源。",
-      "attemptTitle": "Attempt",
-      "batchProgress": "共 {{total}} 个 · {{queued}} 个排队 · {{running}} 个运行中 · {{launched}} 个已启动 · {{failed}} 个失败 · {{cancelled}} 个已取消",
-      "sessionGroupTitle": "Benchmark sessions",
-      "sessionGroupProgress": "共 {{total}} 个 · {{queued}} 个排队 · {{running}} 个运行中 · {{launched}} 个已启动 · {{failed}} 个失败 · {{cancelled}} 个已取消",
-      "evaluateSubmitted": "评估",
-      "evaluatingSubmitted": "正在评估补丁...",
-      "taskActionPlaceholder": "Task IDs for add/remove/stop/restart",
-      "sessionLabel": "Session：{{sessionId}}",
-      "noSessionYet": "暂无 Session",
-      "testResultPending": "测试结果待定",
-      "taskBreadcrumb": "{{attempt}} > {{task}}"
-    },
     "skillsAndTools": "Skills 与工具",
     "agentSpotlightViewMode": "Agent 视图模式",
     "agentSpotlightListView": "列表视图",
diff --git a/src/modules/WorkStation/AppShell/CodeSidebarHeaderActions.tsx b/src/modules/WorkStation/AppShell/CodeSidebarHeaderActions.tsx
index dc36eb8ce4..1b4bdf6733 100644
--- a/src/modules/WorkStation/AppShell/CodeSidebarHeaderActions.tsx
+++ b/src/modules/WorkStation/AppShell/CodeSidebarHeaderActions.tsx
@@ -44,7 +44,6 @@ function usesFallbackCodeSidebar(tab: WorkStationTab | null): boolean {
     tab.type !== "github-pr-detail" &&
     tab.type !== "source-control" &&
     tab.type !== "terminal" &&
-    tab.type !== "benchmark" &&
     tab.type !== "search-sessions"
   );
 }
diff --git a/src/modules/WorkStation/TabContent/registry.ts b/src/modules/WorkStation/TabContent/registry.ts
index ef3a6f5874..b9f4322baa 100644
--- a/src/modules/WorkStation/TabContent/registry.ts
+++ b/src/modules/WorkStation/TabContent/registry.ts
@@ -208,15 +208,6 @@ const WorkItemDetailEntry: RendererEntry = {
   debugLabel: "workItem-detail",
 };
 
-// ============================================
-// Launchpad renderers
-// ============================================
-
-const BenchmarkEntry: RendererEntry = {
-  Component: lazy(() => import("./renderers/benchmark")),
-  debugLabel: "benchmark",
-};
-
 // ============================================
 // Canvas Preview renderer
 // ============================================
@@ -269,7 +260,6 @@ export const REGISTRY: TabContentRegistry = {
   "lint-scan": LintScanEntry,
   "ai-impact": AIImpactEntry,
   "search-sessions": SearchSessionsEntry,
-  benchmark: BenchmarkEntry,
   "url-preview": UrlPreviewEntry,
 
   // Browser
diff --git a/src/modules/WorkStation/shared/SidebarModules/index.ts b/src/modules/WorkStation/shared/SidebarModules/index.ts
index 4c71af8627..a6b9ed459a 100644
--- a/src/modules/WorkStation/shared/SidebarModules/index.ts
+++ b/src/modules/WorkStation/shared/SidebarModules/index.ts
@@ -20,8 +20,6 @@ export {
 
 export { TerminalTabSidebar } from "./Terminal";
 
-export { BenchmarkTabSidebar } from "./Benchmark";
-
 export {
   registerTabSidebar,
   getTabSidebarDescriptor,
diff --git a/src/modules/WorkStation/shared/TabBar/components/SortableTab/index.tsx b/src/modules/WorkStation/shared/TabBar/components/SortableTab/index.tsx
index 94fc5c2aa3..414b9fef83 100644
--- a/src/modules/WorkStation/shared/TabBar/components/SortableTab/index.tsx
+++ b/src/modules/WorkStation/shared/TabBar/components/SortableTab/index.tsx
@@ -8,7 +8,6 @@ import { useSortable } from "@dnd-kit/sortable";
 import { useAtomValue } from "jotai";
 import {
   Infinity,
-  BookLock,
   Box,
   Building2,
   CircleDot,
@@ -83,7 +82,6 @@ import { WorkStationTabPillSurface } from "../WorkStationTabPillSurface";
 // ============================================
 
 const WORKSTATION_TAB_ICONS = {
-  BookLock,
   Box,
   Building2,
   CircleDot,
@@ -236,16 +234,6 @@ export const SortableTab: React.FC<SortableTabProps> = memo(
         );
       }
 
-      if (tab.type === "benchmark") {
-        return (
-          <BookLock
-            size={16}
-            strokeWidth={1.75}
-            className={isActive ? "text-primary-6" : "text-text-2"}
-          />
-        );
-      }
-
       if (tab.type === "chat-session") {
         return (
           <ChatSessionTabIcon
diff --git a/src/scaffold/NavigationSidebar/connectors/WorkstationSidebarConnector/sidebarConnector.pinnedAndRevealData.ts b/src/scaffold/NavigationSidebar/connectors/WorkstationSidebarConnector/sidebarConnector.pinnedAndRevealData.ts
index 232d8ac2cb..dd3ef7eca0 100644
--- a/src/scaffold/NavigationSidebar/connectors/WorkstationSidebarConnector/sidebarConnector.pinnedAndRevealData.ts
+++ b/src/scaffold/NavigationSidebar/connectors/WorkstationSidebarConnector/sidebarConnector.pinnedAndRevealData.ts
@@ -1,6 +1,6 @@
 /**
  * Pinned menu items, the rename modal, the currently-highlighted session id
- * (chat-panel terminal tab / benchmark master row / active session), and
+ * (chat-panel terminal tab / active session), and
  * the merged reveal-candidate list for `WorkstationSidebarConnector`
  * (`index.tsx`).
  */
@@ -9,7 +9,6 @@ import { useAtomValue } from "jotai";
 import { useMemo } from "react";
 
 import type { NavigationMenuItem } from "@src/scaffold/NavigationSidebar/components/NavigationMenu/config";
-import { benchmarkAgentBatchStatusAtom } from "@src/store/benchmark";
 import { activeChatPanelTabAtom } from "@src/store/chatPanel/chatPanelTabsAtom";
 import type { SessionCreatorDraft } from "@src/store/session";
 import { toChatPanelTuiSessionId } from "@src/util/ui/terminal/chatPanelTuiSessionId";
@@ -58,18 +57,13 @@ export function useWorkstationSidebarPinnedAndRevealData({
 }: UseWorkstationSidebarPinnedAndRevealDataParams) {
   const rename = useRenameSessionModal();
   const activeChatPanelTab = useAtomValue(activeChatPanelTabAtom);
-  const benchmarkBatchStatus = useAtomValue(benchmarkAgentBatchStatusAtom);
   const activeChatPanelTuiSessionId =
     activeChatPanelTab?.type === "terminal"
       ? toChatPanelTuiSessionId(activeChatPanelTab.id)
       : "";
   const highlightedSessionId = activeChatPanelTuiSessionId
     ? activeChatPanelTuiSessionId
-    : benchmarkBatchStatus?.items.some(
-          (item) => item.sessionId === activeSessionId
-        )
-      ? benchmarkBatchStatus.masterSessionId
-      : activeSessionId;
+    : activeSessionId;
 
   const workItemsSidebarMenuItems = useMemo(
     () =>
diff --git a/src/scaffold/NavigationSidebar/connectors/useSessionMenuItems/index.tsx b/src/scaffold/NavigationSidebar/connectors/useSessionMenuItems/index.tsx
index 2c4bffb322..0c71092117 100644
--- a/src/scaffold/NavigationSidebar/connectors/useSessionMenuItems/index.tsx
+++ b/src/scaffold/NavigationSidebar/connectors/useSessionMenuItems/index.tsx
@@ -3,12 +3,9 @@ import { useAtomValue } from "jotai";
 import { useCallback, useEffect, useMemo, useState } from "react";
 import { useTranslation } from "react-i18next";
 
-import { benchmarkApi } from "@src/api/tauri/benchmark";
 import type { AgentLiveStatus } from "@src/api/tauri/rpc/schemas/agentOrgs";
-import { createLogger } from "@src/hooks/logger";
 import { useFilteredItems } from "@src/hooks/search";
 import type { NavigationMenuItem } from "@src/scaffold/NavigationSidebar/components/NavigationMenu/config";
-import { benchmarkAgentBatchStatusAtom } from "@src/store/benchmark";
 import {
   type Session,
   type SessionListCategory,
@@ -29,11 +26,7 @@ import {
   DEFAULT_GROUP_VISIBLE_COUNT,
   type DateGroupKey,
 } from "./dateGroupingHelpers";
-import {
-  buildSessionMenuItem,
-  isBenchmarkSessionRow,
-  separator,
-} from "./menuItemBuilders";
+import { buildSessionMenuItem, separator } from "./menuItemBuilders";
 import {
   buildByAgentMenuItems,
   buildByTimeMenuItems,
@@ -71,8 +64,6 @@ function liveDetailForSession(
 
 export { getLoadMoreGroupId, isLoadMoreId } from "./paginationHelpers";
 
-const logger = createLogger("SessionSidebar");
-
 interface ChildSessionRecord {
   sessionId: string;
   name: string;
@@ -186,9 +177,6 @@ export function useSessionMenuItems({
   const { t: tCommon } = useTranslation();
   const pagination = useAtomValue(sessionPaginationAtom);
   const agentLiveStatuses = useAtomValue(agentLiveStatusAtom);
-  const benchmarkAgentBatchStatus = useAtomValue(benchmarkAgentBatchStatusAtom);
-  const [benchmarkHistoryChildSessionIds, setBenchmarkHistoryChildSessionIds] =
-    useState<ReadonlySet<string>>(() => new Set());
   // parentId → the parent's updated_at at query time. Children are re-fetched
   // only when the parent session changes, instead of re-querying every
   // visible session on every list refresh (that pattern issued 100+
@@ -199,50 +187,6 @@ export function useSessionMenuItems({
   const [fetchedChildSessionsByParent, setFetchedChildSessionsByParent] =
     useState<ReadonlyMap<string, Session[]>>(() => new Map());
 
-  useEffect(() => {
-    let cancelled = false;
-    benchmarkApi
-      .listAgentBatchHistories()
-      .then((histories) => {
-        if (cancelled) return;
-        setBenchmarkHistoryChildSessionIds(
-          new Set(
-            histories.flatMap((history) =>
-              history.items
-                .map((item) => item.sessionId)
-                .filter((sessionId): sessionId is string => Boolean(sessionId))
-            )
-          )
-        );
-      })
-      .catch((error: unknown) => {
-        if (cancelled) return;
-        logger.warn("Failed to load benchmark batch histories:", error);
-      });
-    return () => {
-      cancelled = true;
-    };
-  }, []);
-
-  const benchmarkChildSessionIds = useMemo(
-    () =>
-      new Set(
-        benchmarkAgentBatchStatus?.items
-          .map((item) => item.sessionId)
-          .filter((sessionId): sessionId is string => Boolean(sessionId)) ?? []
-      ),
-    [benchmarkAgentBatchStatus?.items]
-  );
-
-  const benchmarkCoordinatorSessionIds = useMemo(
-    () =>
-      new Set(
-        sortedSessions
-          .filter(isBenchmarkSessionRow)
-          .map((session) => session.session_id)
-      ),
-    [sortedSessions]
-  );
   const isInSidebarRoster = useMemo(
     () => createSidebarRosterMatcher(pagination),
     [pagination]
@@ -273,16 +217,10 @@ export function useSessionMenuItems({
               (includeExternal ||
                 !isImportedHistorySession(session.session_id)) &&
               (sessionMatchesOrgFilter(session, selectedOrgIds) ||
-                (extraSessionIds?.has(session.session_id) ?? false)))) &&
-          !benchmarkChildSessionIds.has(session.session_id) &&
-          !benchmarkHistoryChildSessionIds.has(session.session_id) &&
-          !benchmarkCoordinatorSessionIds.has(session.parentSessionId ?? "")
+                (extraSessionIds?.has(session.session_id) ?? false))))
         );
       }),
     [
-      benchmarkChildSessionIds,
-      benchmarkCoordinatorSessionIds,
-      benchmarkHistoryChildSessionIds,
       extraSessionIds,
       includeExternal,
       isInSidebarRoster,
diff --git a/src/scaffold/NavigationSidebar/connectors/useSessionMenuItems/menuItemBuilders.tsx b/src/scaffold/NavigationSidebar/connectors/useSessionMenuItems/menuItemBuilders.tsx
index 36488d527c..bb3919886d 100644
--- a/src/scaffold/NavigationSidebar/connectors/useSessionMenuItems/menuItemBuilders.tsx
+++ b/src/scaffold/NavigationSidebar/connectors/useSessionMenuItems/menuItemBuilders.tsx
@@ -30,10 +30,6 @@ export {
   isSessionPendingAsking,
 } from "@src/util/session/sessionStatusDot";
 
-export function isBenchmarkSessionRow(session: Session): boolean {
-  return session.user_input?.startsWith("Benchmark run coordinator") ?? false;
-}
-
 interface BuildSessionMenuItemParams {
   session: Session;
   untitledSession: string;
diff --git a/src/scaffold/NavigationSidebar/connectors/useWorkstationSidebarHandlers.ts b/src/scaffold/NavigationSidebar/connectors/useWorkstationSidebarHandlers.ts
index 34248d73a2..8cec559086 100644
--- a/src/scaffold/NavigationSidebar/connectors/useWorkstationSidebarHandlers.ts
+++ b/src/scaffold/NavigationSidebar/connectors/useWorkstationSidebarHandlers.ts
@@ -4,7 +4,6 @@ import { useAtomValue, useSetAtom } from "jotai";
 import { type Dispatch, type SetStateAction, useCallback } from "react";
 
 import { deleteSession } from "@src/api/tauri/agent";
-import { benchmarkApi } from "@src/api/tauri/benchmark";
 import { deleteHumanSession } from "@src/api/tauri/humanSession";
 import { rpc } from "@src/api/tauri/rpc";
 import Message from "@src/components/Message";
@@ -35,11 +34,6 @@ import { clearCliTurnLifecycleSession } from "@src/hooks/cliSession/cliTurnLifec
 import { createLogger } from "@src/hooks/logger";
 import type { GoToNewSessionOptions } from "@src/hooks/navigation/useAppNavigation";
 import type { NavigationMenuItem } from "@src/scaffold/NavigationSidebar/components/NavigationMenu/config";
-import {
-  benchmarkActiveBatchIdAtom,
-  benchmarkActiveBatchTaskIdAtom,
-  benchmarkAgentBatchStatusAtom,
-} from "@src/store/benchmark";
 import {
   SESSION_SIDEBAR_PAGE_SIZE,
   type Session,
@@ -141,13 +135,6 @@ export function useWorkstationSidebarHandlers({
   onCloudSidebarItemClick,
 }: UseWorkstationSidebarHandlersParams): UseWorkstationSidebarHandlersResult {
   const navigateChatPanel = useSetAtom(chatPanelNavigateAtom);
-  const setBenchmarkAgentBatchStatus = useSetAtom(
-    benchmarkAgentBatchStatusAtom
-  );
-  const setBenchmarkActiveBatchId = useSetAtom(benchmarkActiveBatchIdAtom);
-  const setBenchmarkActiveBatchTaskId = useSetAtom(
-    benchmarkActiveBatchTaskIdAtom
-  );
   const disposeWorkstationWorkspace = useSetAtom(
     disposeWorkstationWorkspaceAtom
   );
@@ -373,26 +360,6 @@ export function useWorkstationSidebarHandlers({
         sessionRouteLabel
       );
 
-      if (isBenchmarkCoordinatorSession(originalSession)) {
-        navigateChatPanel({
-          kind: CHAT_PANEL_SURFACE_KIND.BENCHMARK_SESSION_GROUP,
-        });
-        promoteActiveSessionCreatorDraft();
-        void benchmarkApi
-          .listAgentBatchHistories({ limit: 100 })
-          .then((histories) =>
-            histories.find((history) => history.masterSessionId === item.id)
-          )
-          .then((history) => {
-            if (!history) return;
-            setBenchmarkAgentBatchStatus(history);
-            setBenchmarkActiveBatchId(history.batchId);
-            setBenchmarkActiveBatchTaskId(null);
-          });
-        openSession(item.id, sessionName, originalSession.repoPath);
-        return;
-      }
-
       navigateChatPanel({ kind: CHAT_PANEL_SURFACE_KIND.SESSION });
       promoteActiveSessionCreatorDraft();
       onOpenSessionChatPanelTab({
@@ -417,9 +384,6 @@ export function useWorkstationSidebarHandlers({
       onOpenSessionChatPanelTab,
       promoteActiveSessionCreatorDraft,
       sessionRouteLabel,
-      setBenchmarkActiveBatchId,
-      setBenchmarkActiveBatchTaskId,
-      setBenchmarkAgentBatchStatus,
       setGroupVisibleCounts,
     ]
   );
@@ -464,7 +428,3 @@ function loadMoreCategoryAction(
 ): ReturnType<typeof loadMoreCategory> {
   return loadMoreCategory(sessionListCategory);
 }
-
-function isBenchmarkCoordinatorSession(session: Session): boolean {
-  return session.user_input?.startsWith("Benchmark run coordinator\n") ?? false;
-}
diff --git a/src/store/ui/chatPanelAtom.ts b/src/store/ui/chatPanelAtom.ts
index a4d1096731..10bd0c7d97 100644
--- a/src/store/ui/chatPanelAtom.ts
+++ b/src/store/ui/chatPanelAtom.ts
@@ -277,7 +277,6 @@ export const CHAT_PANEL_CREATE_TARGET = {
   GITHUB_ISSUES_PROJECT: "githubIssuesProject",
   WORK_ITEM: "workItem",
   COLLAB_ORG: "collabOrg",
-  BENCHMARK: "benchmark",
 } as const;
 
 export type ChatPanelCreateTarget =
@@ -339,7 +338,6 @@ chatPanelCreateProjectContextAtom.debugLabel =
 export const CHAT_PANEL_CONTENT_MODE = {
   SESSION: "session",
   NON_SESSION: "nonSession",
-  BENCHMARK_SESSION_GROUP: "benchmarkSessionGroup",
 } as const;
 
 export type ChatPanelContentMode =
@@ -482,7 +480,6 @@ chatPanelWorkspaceOverviewTabAtom.debugLabel =
 
 export type ChatPanelSurfaceState =
   | { kind: typeof CHAT_PANEL_SURFACE_KIND.SESSION }
-  | { kind: typeof CHAT_PANEL_SURFACE_KIND.BENCHMARK_SESSION_GROUP }
   | { kind: typeof CHAT_PANEL_SURFACE_KIND.NEW_PROJECT }
   | { kind: typeof CHAT_PANEL_SURFACE_KIND.NEW_GITHUB_ISSUES_PROJECT }
   | { kind: typeof CHAT_PANEL_SURFACE_KIND.NEW_WORK_ITEM }
@@ -512,7 +509,6 @@ export type ChatPanelSurfaceState =
 
 export type ChatPanelNavigateCommand =
   | { kind: typeof CHAT_PANEL_SURFACE_KIND.SESSION }
-  | { kind: typeof CHAT_PANEL_SURFACE_KIND.BENCHMARK_SESSION_GROUP }
   | {
       kind: typeof CHAT_PANEL_SURFACE_KIND.NEW_PROJECT;
       createProjectContext?: ChatPanelCreateProjectContext | null;
@@ -578,12 +574,6 @@ export const chatPanelNavigateAtom = atom(
       case CHAT_PANEL_SURFACE_KIND.SESSION:
         set(chatPanelContentModeAtom, CHAT_PANEL_CONTENT_MODE.SESSION);
         return;
-      case CHAT_PANEL_SURFACE_KIND.BENCHMARK_SESSION_GROUP:
-        set(
-          chatPanelContentModeAtom,
-          CHAT_PANEL_CONTENT_MODE.BENCHMARK_SESSION_GROUP
-        );
-        return;
       case CHAT_PANEL_SURFACE_KIND.NEW_PROJECT:
         set(chatPanelContentModeAtom, CHAT_PANEL_CONTENT_MODE.NON_SESSION);
         set(chatPanelCreateTargetAtom, CHAT_PANEL_CREATE_TARGET.PROJECT);
@@ -649,11 +639,6 @@ export const chatPanelNavigateAtom = atom(
 chatPanelNavigateAtom.debugLabel = "chatPanelNavigateAtom";
 
 export const activeChatPanelSurfaceAtom = atom<ChatPanelSurfaceState>((get) => {
-  const contentMode = get(chatPanelContentModeAtom);
-  if (contentMode === CHAT_PANEL_CONTENT_MODE.BENCHMARK_SESSION_GROUP) {
-    return { kind: CHAT_PANEL_SURFACE_KIND.BENCHMARK_SESSION_GROUP };
-  }
-
   const selectedWorkItem = get(chatPanelSelectedWorkItemAtom);
   if (selectedWorkItem) {
     return {
@@ -696,6 +681,7 @@ export const activeChatPanelSurfaceAtom = atom<ChatPanelSurfaceState>((get) => {
     };
   }
 
+  const contentMode = get(chatPanelContentModeAtom);
   const createTarget = get(chatPanelCreateTargetAtom);
   if (
     contentMode === CHAT_PANEL_CONTENT_MODE.NON_SESSION &&
diff --git a/src/store/workstation/tabs/__tests__/workspaceState.test.ts b/src/store/workstation/tabs/__tests__/workspaceState.test.ts
index 023d3f38b7..6be41bdad8 100644
--- a/src/store/workstation/tabs/__tests__/workspaceState.test.ts
+++ b/src/store/workstation/tabs/__tests__/workspaceState.test.ts
@@ -48,7 +48,6 @@ const EXPECTED_OWNERSHIP: Record<WorkStationTabType, WorkstationTabOwnership> =
     "lint-scan": "workspace-local",
     "ai-impact": "workspace-local",
     "search-sessions": "workspace-local",
-    benchmark: "shared-resource",
     "url-preview": "workspace-local",
     "browser-session": "shared-resource",
     devtools: "shared-resource",
@@ -119,7 +118,7 @@ describe("WorkStation tab ownership policy", () => {
       })
     );
 
-    expect(results).toHaveLength(39);
+    expect(results).toHaveLength(38);
     expect(results.every(({ actual, expected }) => actual === expected)).toBe(
       true
     );
diff --git a/src/store/workstation/tabs/factories/codeEditor.ts b/src/store/workstation/tabs/factories/codeEditor.ts
index eb7abebd8b..d8748f40eb 100644
--- a/src/store/workstation/tabs/factories/codeEditor.ts
+++ b/src/store/workstation/tabs/factories/codeEditor.ts
@@ -578,28 +578,6 @@ export function createSearchSessionsTab(): WorkStationTab {
   return searchSessionsTabFactory({});
 }
 
-export interface BenchmarkTabData {
-  batchId?: string;
-  selectedTaskId?: string;
-}
-
-export const benchmarkTabFactory = defineTabFactory<BenchmarkTabData>({
-  tabType: "benchmark",
-  idStrategy: {
-    type: "keyed",
-    prefix: "benchmark",
-    getKey: (data) => data.batchId ?? "main",
-  },
-  getTitle: (data) => (data.batchId ? "Benchmark Run" : "Benchmark"),
-  icon: "BookLock",
-});
-
-export function createBenchmarkTab(
-  data: BenchmarkTabData = {}
-): WorkStationTab {
-  return benchmarkTabFactory(data);
-}
-
 export const lintScanTabFactory = defineTabFactory<{ repoPath: string }>({
   tabType: "lint-scan",
   idStrategy: { type: "singleton", id: "lint-scan:main" },
diff --git a/src/store/workstation/tabs/factories/index.ts b/src/store/workstation/tabs/factories/index.ts
index 2f01edf2ae..d163e115fa 100644
--- a/src/store/workstation/tabs/factories/index.ts
+++ b/src/store/workstation/tabs/factories/index.ts
@@ -24,7 +24,6 @@ export {
   settingsTabFactory,
   aiImpactTabFactory,
   searchSessionsTabFactory,
-  benchmarkTabFactory,
   lintScanTabFactory,
   searchTabFactory,
   SOURCE_CONTROL_CHANGES_TAB_ID,
@@ -48,7 +47,6 @@ export {
   createSettingsTab,
   createAIImpactTab,
   createSearchSessionsTab,
-  createBenchmarkTab,
   createLintScanTab,
   createSearchTab,
   urlPreviewTabFactory,
diff --git a/src/store/workstation/tabs/index.ts b/src/store/workstation/tabs/index.ts
index 6ee356b49d..cf36b54599 100644
--- a/src/store/workstation/tabs/index.ts
+++ b/src/store/workstation/tabs/index.ts
@@ -114,7 +114,6 @@ export {
   settingsTabFactory,
   aiImpactTabFactory,
   searchSessionsTabFactory,
-  benchmarkTabFactory,
   lintScanTabFactory,
   searchTabFactory,
   // Code Editor creator functions
@@ -138,7 +137,6 @@ export {
   createSettingsTab,
   createAIImpactTab,
   createSearchSessionsTab,
-  createBenchmarkTab,
   createLintScanTab,
   createSearchTab,
   // Browser factories
diff --git a/src/store/workstation/tabs/storage.ts b/src/store/workstation/tabs/storage.ts
index 25138cfa76..c7d2cd9d89 100644
--- a/src/store/workstation/tabs/storage.ts
+++ b/src/store/workstation/tabs/storage.ts
@@ -40,7 +40,6 @@ const VALID_WORKSTATION_TAB_TYPES = new Set<WorkStationTabType>([
   "lint-scan",
   "ai-impact",
   "search-sessions",
-  "benchmark",
   "url-preview",
   "browser-session",
   "devtools",
diff --git a/src/store/workstation/tabs/tabFactory.ts b/src/store/workstation/tabs/tabFactory.ts
index aa9b43d3eb..34f806d22a 100644
--- a/src/store/workstation/tabs/tabFactory.ts
+++ b/src/store/workstation/tabs/tabFactory.ts
@@ -88,7 +88,6 @@ const DEFAULT_CATEGORY_BY_TYPE: Record<
   "lint-scan": "lint",
   "ai-impact": "ai-impact",
   "search-sessions": "search-sessions",
-  benchmark: "benchmark",
   "url-preview": "preview",
   "browser-session": "browser",
   devtools: "browser",
diff --git a/src/store/workstation/tabs/types.ts b/src/store/workstation/tabs/types.ts
index 88d578d5fa..41266abefe 100644
--- a/src/store/workstation/tabs/types.ts
+++ b/src/store/workstation/tabs/types.ts
@@ -34,7 +34,6 @@ export type WorkStationTabType =
   | "lint-scan" // Workspace lint scan configuration
   | "ai-impact" // AI session impact dashboard
   | "search-sessions" // Session search + table (reuses SessionTable; launchpad tab)
-  | "benchmark" // Benchmark task browser and runner setup
   | "url-preview" // URL preview (agent-triggered webview in editor)
   // Browser tabs
   | "browser-session"
@@ -97,7 +96,6 @@ export type WorkStationTabCategory =
   | "lint"
   | "ai-impact"
   | "search-sessions"
-  | "benchmark"
   | "preview"
   | "subagent"
   | "agent-config"
@@ -251,7 +249,6 @@ export function getWorkstationTabOwnership(
 
     case "terminal":
     case "settings":
-    case "benchmark":
     case "browser-session":
     case "devtools":
     case "project-dashboard":
diff --git a/src/types/ui/chatPanel.ts b/src/types/ui/chatPanel.ts
index 32856de0bb..ae6ce39a8c 100644
--- a/src/types/ui/chatPanel.ts
+++ b/src/types/ui/chatPanel.ts
@@ -7,7 +7,6 @@
  */
 export const CHAT_PANEL_SURFACE_KIND = {
   SESSION: "session",
-  BENCHMARK_SESSION_GROUP: "benchmarkSessionGroup",
   NEW_PROJECT: "newProject",
   NEW_GITHUB_ISSUES_PROJECT: "newGithubIssuesProject",
   NEW_WORK_ITEM: "newWorkItem",
diff --git a/src/util/session/__tests__/sessionSidebarRow.test.ts b/src/util/session/__tests__/sessionSidebarRow.test.ts
index 679d37a638..238b46626d 100644
--- a/src/util/session/__tests__/sessionSidebarRow.test.ts
+++ b/src/util/session/__tests__/sessionSidebarRow.test.ts
@@ -1,4 +1,3 @@
-import { FlaskConical } from "lucide-react";
 import { describe, expect, it } from "vitest";
 
 import {
@@ -119,15 +118,4 @@ describe("resolveSessionRowIcon", () => {
       }).isMonochromeBrandIcon
     ).toBe(false);
   });
-
-  it("keeps benchmark coordinator sessions on the benchmark icon", () => {
-    expect(
-      resolveSessionRowIcon({
-        session_id: "cliagent-benchmark",
-        user_input: "Benchmark run coordinator for OpenCode",
-        cliAgentType: "opencode",
-        agentIconId: "codex",
-      })
-    ).toBe(FlaskConical);
-  });
 });
diff --git a/src/util/session/sessionDisplayMetadata.ts b/src/util/session/sessionDisplayMetadata.ts
index 8f333a9750..4a8245cb8d 100644
--- a/src/util/session/sessionDisplayMetadata.ts
+++ b/src/util/session/sessionDisplayMetadata.ts
@@ -76,7 +76,6 @@ interface NormalizedSessionDisplayInput {
   modelName?: string;
   externalHistorySource?: string;
   imported: boolean;
-  benchmark: boolean;
   agentOrg: boolean;
   remoteNative: boolean;
 }
@@ -98,7 +97,6 @@ function normalizeSessionDisplayInput(
           ? session.origin.source
           : undefined,
       imported: false,
-      benchmark: false,
       agentOrg: false,
       remoteNative: !session.origin || session.origin.kind === "orgii",
     };
@@ -118,8 +116,6 @@ function normalizeSessionDisplayInput(
     modelName: sourceDisplay?.model ?? session.model,
     externalHistorySource: session.importedFrom?.externalHistorySource,
     imported: Boolean(session.importedFrom),
-    benchmark:
-      session.user_input?.startsWith("Benchmark run coordinator") ?? false,
     agentOrg: Boolean(session.agentOrgId),
     remoteNative: false,
   };
@@ -158,7 +154,6 @@ function resolveAgentIconId(
   agentType: string | undefined,
   externalSource: ImportedHistorySourceDescriptor | undefined
 ): string {
-  if (input.benchmark) return "flask-conical";
   if (input.agentOrg) return "network";
   if (externalSource) return externalSource.iconId;
 
diff --git a/tests/e2e/wdio.conf.mjs b/tests/e2e/wdio.conf.mjs
index a965ebb0be..14b50e5b27 100644
--- a/tests/e2e/wdio.conf.mjs
+++ b/tests/e2e/wdio.conf.mjs
@@ -292,7 +292,7 @@ process.env.ORGII_EXTERNAL_HISTORY_HOME = externalHistoryHome;
 // must exist on disk before the app process launches so the app's own
 // startup external-history auto-scan (`useDataSourceAutoScan`) discovers it
 // without the spec needing a debug seed/mutation endpoint. Written here
-// (mirroring `ensureE2EWorkspaceRepo`/`ensureBenchmarkDockerFixtureRepo`)
+// (mirroring `ensureE2EWorkspaceRepo`)
 // rather than in the spec so it is guaranteed to land before
 // `startTauriWebDriver()` runs below.
 const CLAUDE_IMPORT_FIXTURE_UUID = "e2ec0de0-c0de-4000-8000-000000000001";
@@ -376,64 +376,10 @@ function ensureClaudeCodeImportFixtureTranscript() {
   return fixturePath;
 }
 
-function ensureBenchmarkDockerFixtureRepo() {
-  if (process.env.ORGII_SWE_BENCH_PRO_REPO_PATH) return;
-  const fixtureRoot = join(tmpdir(), "orgii-e2e-swe-bench-pro-fixture");
-  const runScriptsDir = join(fixtureRoot, "run_scripts", "e2e_docker_task");
-  mkdirSync(runScriptsDir, { recursive: true });
-  writeFileSync(
-    join(fixtureRoot, "swe_bench_pro_eval.py"),
-    `#!/usr/bin/env python3
-import argparse
-import json
-import os
-import subprocess
-import sys
-
-parser = argparse.ArgumentParser()
-parser.add_argument("--raw_sample_path", required=True)
-parser.add_argument("--patch_path", required=True)
-parser.add_argument("--output_dir", required=True)
-parser.add_argument("--scripts_dir", required=True)
-parser.add_argument("--dockerhub_username")
-parser.add_argument("--use_local_docker", action="store_true")
-parser.add_argument("--num_workers", default="1")
-args = parser.parse_args()
-
-with open(args.patch_path, "r", encoding="utf-8") as handle:
-    patch_rows = json.load(handle)
-task_id = patch_rows[0]["instance_id"]
-command = ["docker", "run", "--rm", "alpine:3.20", "sh", "-lc", "echo orgii-docker-benchmark-e2e"]
-print("running docker command:", " ".join(command), flush=True)
-completed = subprocess.run(command, text=True, capture_output=True)
-print(completed.stdout, end="", flush=True)
-if completed.stderr:
-    print(completed.stderr, end="", file=sys.stderr, flush=True)
-os.makedirs(args.output_dir, exist_ok=True)
-with open(os.path.join(args.output_dir, "eval_results.json"), "w", encoding="utf-8") as handle:
-    json.dump({task_id: completed.returncode == 0 and "orgii-docker-benchmark-e2e" in completed.stdout}, handle)
-sys.exit(completed.returncode)
-`,
-    "utf8"
-  );
-  writeFileSync(
-    join(runScriptsDir, "run_script.sh"),
-    "#!/usr/bin/env bash\necho e2e run script\n",
-    "utf8"
-  );
-  writeFileSync(
-    join(runScriptsDir, "parser.py"),
-    "print('e2e parser')\n",
-    "utf8"
-  );
-  process.env.ORGII_SWE_BENCH_PRO_REPO_PATH = fixtureRoot;
-}
-
 process.env.ORGII_IDE_SERVER_PORT = String(ideServerPort);
 process.env.E2E_BASE_URL =
   process.env.E2E_BASE_URL ?? `http://127.0.0.1:${ideServerPort}`;
 ensureE2EWorkspaceRepo();
-ensureBenchmarkDockerFixtureRepo();
 ensureClaudeCodeImportFixtureTranscript();
 
 const WDIO_PRE_FLIGHT_PORTS = [webDriverPort, frontendPort, ideServerPort];