Remove Normalized E2E / Session Time / Prefill TPS x-axis modes / 移除 Normalized E2E、会话时长与 Prefill TPS X 轴模式

packages/app/cypress/e2e/ttft-x-axis-toggle.cy.ts

@@ -1,27 +1,5 @@
 import { unlockAgenticGate } from '../support/e2e';
 
-const interceptDerivedMetrics = () => {
-  cy.intercept('GET', '/api/v1/derived-agentic-metrics*', (request) => {
-    const ids = new URL(request.url).searchParams.get('ids')?.split(',').filter(Boolean) ?? [];
-    request.reply({
-      body: Object.fromEntries(
-        ids.map((id, index) => [
-          id,
-          {
-            id: Number(id),
-            normalized_session_time_s: 60 + index,
-            p90_prefill_tps_per_user: 100 + index,
-            p75_normalized_e2e_400_s: 8 + index,
-            p90_normalized_e2e_400_s: 12 + index,
-            p75_osl_per_e2el: 40 + index,
-            p90_osl_per_e2el: 25 + index,
-          },
-        ]),
-      ),
-    });
-  }).as('derivedAgenticMetrics');
-};
-
 // This spec exercises the agentic x-axis modes, which only exist when the
 // selected model resolves to the Agentic Traces scenario. The default e2e
 // fixtures (cypress/fixtures/api/*.json) have NO agentic rows for any model, and
@@ -148,10 +126,6 @@ const interceptFixedSequenceData = () => {
 describe('X-Axis Mode Toggle (inference chart)', () => {
   before(() => {
     interceptAgenticData();
-    // Stub derived metrics before the visit: if the agentic DEFAULT x-axis mode
-    // is (or becomes) a derived mode that fetches /derived-agentic-metrics on
-    // mount, the chart must not sit on its loading skeleton.
-    interceptDerivedMetrics();
     cy.visit('/inference?i_seq=agentic-traces', {
       onBeforeLoad(win) {
         win.localStorage.setItem('inferencex-star-modal-dismissed', String(Date.now()));
@@ -166,7 +140,6 @@ describe('X-Axis Mode Toggle (inference chart)', () => {
     cy.get('[data-testid="scenario-selector"]').should('contain.text', 'Agentic Traces');
     cy.get('[data-testid="x-axis-mode-ttft"]').should('be.visible');
     cy.get('[data-testid="x-axis-mode-e2e"]').should('be.visible');
-    cy.get('[data-testid="x-axis-mode-normalized-e2e"]').should('be.visible');
     cy.get('[data-testid="x-axis-mode-interactivity"]')
       .should('be.visible')
       .and('have.attr', 'aria-selected', 'true');
@@ -255,41 +228,26 @@ describe('X-Axis Mode Toggle (inference chart)', () => {
     cy.get('[data-testid="chart-figure"] svg').should('contain.text', 'P90 End-to-end Latency (s)');
   });
 
-  it('switches to request-level normalized E2E at 400 output tokens', () => {
-    interceptDerivedMetrics();
-    cy.get('[data-testid="x-axis-mode-normalized-e2e"]').click();
-    cy.wait('@derivedAgenticMetrics');
-    cy.get('[data-testid="x-axis-mode-normalized-e2e"]').should(
+  it('switches back to Interactivity', () => {
+    cy.get('[data-testid="x-axis-mode-interactivity"]').click();
+    cy.get('[data-testid="x-axis-mode-interactivity"]').should(
       'have.attr',
       'aria-selected',
       'true',
     );
-    cy.get('[data-testid="chart-figure"] h2').should(
-      'contain.text',
-      'P90 Normalized E2E @ 400 output tokens',
-    );
+    cy.get('[data-testid="chart-figure"] h2').should('contain.text', 'Interactivity');
     cy.get('[data-testid="chart-figure"] svg').should(
       'contain.text',
-      'P90 Normalized E2E @ 400 output tokens (s)',
-    );
-
-    cy.get('[data-testid="percentile-selector"]').click();
-    cy.contains('[role="option"]', 'p75').click();
-    cy.get('[data-testid="chart-figure"] h2').should(
-      'contain.text',
-      'P75 Normalized E2E @ 400 output tokens',
+      'P90 Interactivity (tok/s/user)',
     );
   });
 
-  it('switches back to Interactivity', () => {
+  it('follows the percentile selector in the Interactivity axis label', () => {
+    // Select p75 here rather than inheriting it from another test — the axis
+    // label must track the selector on its own.
     cy.get('[data-testid="x-axis-mode-interactivity"]').click();
-    cy.get('[data-testid="x-axis-mode-interactivity"]').should(
-      'have.attr',
-      'aria-selected',
-      'true',
-    );
-    cy.get('[data-testid="chart-figure"] h2').should('contain.text', 'Interactivity');
-    // Percentile was switched to p75 in the previous test — the axis label follows.
+    cy.get('[data-testid="percentile-selector"]').click();
+    cy.contains('[role="option"]', 'p75').click();
     cy.get('[data-testid="chart-figure"] svg').should(
       'contain.text',
       'P75 Interactivity (tok/s/user)',
@@ -411,9 +369,6 @@ const interceptAgenticDataWithOverlay = () => {
 describe('X-Axis Mode Toggle — overlay path (finding #8 regression guard)', () => {
   before(() => {
     interceptAgenticDataWithOverlay();
-    // Same as the main suite: keep the visit independent of the agentic default
-    // x-axis mode (a derived default fetches /derived-agentic-metrics on mount).
-    interceptDerivedMetrics();
     cy.visit(`/inference?unofficialrun=${OVERLAY_RUN_ID}&i_seq=agentic-traces`, {
       onBeforeLoad(win) {
         win.localStorage.setItem('inferencex-star-modal-dismissed', String(Date.now()));
@@ -462,19 +417,4 @@ describe('X-Axis Mode Toggle — overlay path (finding #8 regression guard)', ()
       expect(total).to.be.greaterThan(0);
     });
   });
-
-  it('normalized-e2e mode shows suppression banner for unofficial-run overlays', () => {
-    interceptDerivedMetrics();
-    cy.get('[data-testid="x-axis-mode-normalized-e2e"]').click();
-    cy.get('[data-testid="x-axis-mode-normalized-e2e"]').should(
-      'have.attr',
-      'aria-selected',
-      'true',
-    );
-    // The suppression message appears because isUnofficialRun is true and the
-    // mode is 'normalized-e2e' (documented in ChartDisplay.tsx ~line 640).
-    cy.contains(
-      'Normalized E2E requires persisted per-request traces, so unofficial-run overlays are unavailable for this experimental view.',
-    ).should('be.visible');
-  });
 });

packages/app/src/app/api/v1/agentic-cache-keys.test.ts

@@ -28,17 +28,12 @@ import { STATS_VERSION } from '@semianalysisai/inferencex-db/queries/agentic-agg
 import { REQUEST_TIMELINE_VERSION } from '@semianalysisai/inferencex-db/etl/compute-request-timeline';
 import { TRACE_SERVER_METRICS_VERSION } from '@semianalysisai/inferencex-db/queries/trace-server-metrics';
 
-import { CACHE_KEY_PREFIX as derivedAgenticMetricsKey } from './derived-agentic-metrics/route';
 import { CACHE_KEY_PREFIX as agenticAggregatesKey } from './agentic-aggregates/route';
 import { CACHE_KEY_PREFIX as requestTimelineKey } from './request-timeline/route';
 import { CACHE_KEY_PREFIX as traceServerMetricsKey } from './trace-server-metrics/route';
 import { CACHE_KEY_PREFIX as traceHistogramsKey } from './trace-histograms/route';
 
 describe('agentic blob-cache keys are version-derived', () => {
-  it('derived-agentic-metrics key embeds STATS_VERSION', () => {
-    expect(derivedAgenticMetricsKey).toBe(`derived-agentic-metrics-v${STATS_VERSION}`);
-  });
-
   it('agentic-aggregates key embeds STATS_VERSION', () => {
     expect(agenticAggregatesKey).toBe(`agentic-aggregates-v${STATS_VERSION}`);
   });
@@ -57,7 +52,6 @@ describe('agentic blob-cache keys are version-derived', () => {
 
   it('every key actually contains a version segment (no unversioned literals)', () => {
     for (const key of [
-      derivedAgenticMetricsKey,
       agenticAggregatesKey,
       requestTimelineKey,
       traceServerMetricsKey,

packages/app/src/components/inference/InferenceContext.tsx

@@ -60,12 +60,7 @@ import {
 } from '@/lib/exclusion';
 import { filterRunsByModel, getDisplayLabel } from '@/lib/utils';
 
-import {
-  isAgenticOnlyXAxisMode,
-  useChartData,
-  X_AXIS_MODES,
-  type XAxisMode,
-} from './hooks/useChartData';
+import { useChartData, X_AXIS_MODES, type XAxisMode } from './hooks/useChartData';
 import { resolveComparisonEntries } from './utils/comparisonEntry';
 import {
   comparisonDefaultGroup,
@@ -610,40 +605,23 @@ export function InferenceProvider({
   }, [labelScenarioKind, sequenceResolved, getUrlParam]);
 
   // Reconcile the x-axis mode with the scenario kind:
-  //  - On mount with no `i_xmode` URL param: snap to the kind's natural default
-  //    (interactivity for both agentic and fixed-sequence scenarios). The state was initialized
-  //    to a SSR-stable constant so server and client render the same DOM; this
-  //    effect fixes it up after hydration.
-  //  - When the user later switches sequence kinds: snap to the new kind's
-  //    natural default (the prior selection was for a different kind, so it
-  //    doesn't carry over).
+  //  - On mount with no `i_xmode` URL param: snap to the natural default
+  //    (interactivity for both agentic and fixed-sequence scenarios). The state
+  //    was initialized to a SSR-stable constant so server and client render the
+  //    same DOM; this effect fixes it up after hydration.
+  //  - When the user later switches sequence kinds: snap back to that default
+  //    (the prior selection was for a different kind, so it doesn't carry over).
   const lastSeqKindRef = useRef<ReturnType<typeof sequenceKind> | null>(null);
   useEffect(() => {
     const kind = sequenceKind(effectiveSequence);
     const isInitialMount = lastSeqKindRef.current === null;
-    const isAgenticOnlyMode = isAgenticOnlyXAxisMode(selectedXAxisMode);
-    // On a stale render where kind hasn't changed, bail unless the current
-    // mode is agentic-only and we just landed on a fixed-seq scenario — in
-    // that case force the snap so the chart doesn't try to plot trace-derived
-    // metrics against rows that have no trace_replay.
-    if (!isInitialMount && lastSeqKindRef.current === kind) {
-      if (kind === 'fixed-seq' && isAgenticOnlyMode) {
-        handleSetXAxisMode('interactivity');
-      }
-      return;
-    }
+    // Stale render where the kind hasn't changed — nothing to reconcile.
+    if (!isInitialMount && lastSeqKindRef.current === kind) return;
     lastSeqKindRef.current = kind;
-    if (
-      isInitialMount &&
-      xAxisModeFromUrlRef.current &&
-      !(kind === 'fixed-seq' && isAgenticOnlyMode)
-    ) {
-      // URL-restored agentic-only mode on a fixed-seq sequence makes no sense
-      // — fall through to the default snap below.
-      return;
-    }
+    // A URL-restored mode wins on first mount; later kind switches reset it.
+    if (isInitialMount && xAxisModeFromUrlRef.current) return;
     handleSetXAxisMode('interactivity');
-  }, [effectiveSequence, selectedXAxisMode, handleSetXAxisMode]);
+  }, [effectiveSequence, handleSetXAxisMode]);
 
   // Reconcile selectedE2eXAxisMetric whenever the mode, sequence kind, or
   // agentic percentile changes. For fixed-seq the JSONB only carries

packages/app/src/components/inference/hooks/useChartData.ts

@@ -39,41 +39,18 @@ import {
 /**
  * Chart x-axis variant selected by the mode buttons above the plot. This is
  * the single definition — InferenceContext (URL/state) and ChartDisplay
- * (buttons, derived-metric remapping) import it from here.
+ * (buttons) import it from here.
  */
-export type XAxisMode =
-  | 'ttft'
-  | 'e2e'
-  | 'normalized-e2e'
-  | 'interactivity'
-  | 'session-time'
-  | 'prefill-tps';
-
-export const X_AXIS_MODES: readonly XAxisMode[] = [
-  'ttft',
-  'e2e',
-  'normalized-e2e',
-  'interactivity',
-  'session-time',
-  'prefill-tps',
-];
+export type XAxisMode = 'ttft' | 'e2e' | 'interactivity';
 
-/**
- * Modes whose x metric is derived from persisted per-request traces —
- * these only exist for agentic scenarios (fixed-seq rows have no
- * trace_replay blob to derive them from).
- */
-export function isAgenticOnlyXAxisMode(mode: XAxisMode): boolean {
-  return mode === 'normalized-e2e' || mode === 'session-time' || mode === 'prefill-tps';
-}
+export const X_AXIS_MODES: readonly XAxisMode[] = ['ttft', 'e2e', 'interactivity'];
 
 /**
  * Compute the set of benchmark_results.id values that sit on the
  * (e2e_latency, y) Pareto frontier within each (hwKey, precision, date)
- * group. Used to restrict the non-e2e xmode charts (ttft, interactivity,
- * session-time, prefill-tps) so they show *only* the points that win on
- * end-to-end latency — preventing benchmark-hacking where a config tops
- * one axis while tanking the other.
+ * group. Used to restrict the non-e2e xmode charts (ttft, interactivity) so
+ * they show *only* the points that win on end-to-end latency — preventing
+ * benchmark-hacking where a config tops one axis while tanking the other.
  *
  * Returns null when the y-metric has no roofline direction declared on
  * the e2e chart (caller falls back to no filtering in that case).
@@ -220,7 +197,7 @@ export function useChartData(
   /**
    * Current x-axis mode. When set to anything other than 'e2e', the displayed
    * data is filtered to the (e2e-latency, y) Pareto frontier so the ttft /
-   * interactivity / session-time / prefill-tps charts show only points that
+   * interactivity charts show only points that
    * also win on end-to-end latency — preventing benchmark-hacking where a
    * config tops one metric while tanking the other. The 'e2e' mode is the
    * source of truth and keeps the full point set.
@@ -529,7 +506,7 @@ export function useChartData(
         // non-optimal configs from view.
         //
         // Fixed-seq workloads keep the existing per-axis Pareto since
-        // there's no separate "session-time" notion of total latency —
+        // there's no separate session-level notion of total latency —
         // their e2e IS the request latency, so a TTFT hack there reads
         // honestly on e2e too. The anti-hack constraint is specifically
         // about multi-turn agentic where TTFT measures a tiny fraction

packages/app/src/components/inference/types.ts

@@ -258,7 +258,7 @@ export interface InferenceData extends Partial<Omit<AggDataEntry, AggDataConflic
   /**
    * Whether this point sits on the (e2e_latency, y-metric) Pareto frontier.
    * Set by useChartData when `selectedXAxisMode !== 'e2e'`. The TTFT /
-   * interactivity / session-time / prefill-tps charts use this flag to
+   * interactivity charts use this flag to
    * restrict their roofline computation to e2e-Pareto winners — vendors
    * can't benchmark-hack TTFT by tanking decode (or vice versa) and still
    * appear on the frontier line — while keeping every point visible as
@@ -763,21 +763,10 @@ export interface InferenceChartContextType {
    * at a time, picked by the big buttons above the chart.
    * - 'ttft'          → e2e chartType with x-axis forced to p90_ttft
    * - 'e2e'           → e2e chartType with the chart-config default x-axis (median_e2el / p90_e2el)
-   * - 'normalized-e2e'→ agentic-only; x = per-request E2E normalized to 400 output tokens
    * - 'interactivity' → interactivity chartType (x = median_intvty / p90_intvty)
-   * - 'session-time'  → agentic-only; x = mean-normalized session time (live-computed from trace blobs)
-   * - 'prefill-tps'   → agentic-only; x = mean of P90 prefill TPS/user per session
    */
-  selectedXAxisMode:
-    | 'ttft'
-    | 'e2e'
-    | 'normalized-e2e'
-    | 'interactivity'
-    | 'session-time'
-    | 'prefill-tps';
-  setSelectedXAxisMode: (
-    mode: 'ttft' | 'e2e' | 'normalized-e2e' | 'interactivity' | 'session-time' | 'prefill-tps',
-  ) => void;
+  selectedXAxisMode: 'ttft' | 'e2e' | 'interactivity';
+  setSelectedXAxisMode: (mode: 'ttft' | 'e2e' | 'interactivity') => void;
   scaleType: 'auto' | 'linear' | 'log';
   setScaleType: (type: 'auto' | 'linear' | 'log') => void;
   /** Coarse vendor / framework / agg-disagg / mtp-stp filters applied to the chart point set. */