log scale

.gitignore

@@ -176,8 +176,4 @@ cython_debug/
 evaluation_results/
 images/
 hf_cache/
-*.lock
-
-# macOS
-.DS_Store
-
+*.lock

README.md

@@ -30,6 +30,8 @@ pip install "gradio==5.19.0" pandas -r requirements.txt
 python app.py
 ```
 
+The app is served at `http://127.0.0.1:7860`.
+
 `requirements.txt` lists Plotly. Gradio and pandas are required locally;
 Hugging Face Spaces installs Gradio from the YAML `sdk_version` above.
 

app.py

@@ -62,9 +62,6 @@ custom_css = """
     --pruna-accordion-bg: rgba(255, 255, 255, 0.02);
     --pruna-accordion-border: rgba(216, 180, 254, 0.15);
     --pruna-dropdown-hover: #2a1844;
-    --pruna-toggle-track: var(--pruna-bg-header);
-    --pruna-toggle-thumb: var(--pruna-bg-elevated);
-    --pruna-toggle-thumb-shadow: 0 1px 2px rgba(0, 0, 0, 0.45), inset 0 1px rgba(255, 255, 255, 0.06);
     color-scheme: dark;
 }
 
@@ -108,27 +105,13 @@ custom_css = """
     --pruna-accordion-bg: var(--pruna-bg-card);
     --pruna-accordion-border: var(--pruna-border);
     --pruna-dropdown-hover: #f3e8ff;
-    --pruna-toggle-track: var(--pruna-bg-header);
-    --pruna-toggle-thumb: var(--pruna-bg-card);
-    --pruna-toggle-thumb-shadow: 0 1px 2px rgba(88, 28, 135, 0.12);
     color-scheme: light;
 }
 
-html {
+html, body {
     width: 100% !important;
     max-width: 100% !important;
-    min-width: 0 !important;
-    overflow-x: hidden;
-    overflow-y: auto;
-    -webkit-text-size-adjust: 100%;
-    text-size-adjust: 100%;
-    -webkit-tap-highlight-color: transparent;
-}
-body, gradio-app {
-    width: 100% !important;
-    max-width: 100% !important;
-    min-width: 0 !important;
-    overflow: visible;
+    overflow-x: clip;
     -webkit-tap-highlight-color: transparent;
 }
 html, body, .gradio-container, .main {
@@ -151,25 +134,18 @@ button, a, label, input, select, textarea,
 /* Subtle depth — not a marketing-site hero glow */
 body, .gradio-container {
     background-image: var(--pruna-glow) !important;
-    background-repeat: no-repeat !important;
-    background-attachment: scroll !important;
-}
-@media (min-width: 701px) and (hover: hover) and (pointer: fine) {
-    body, .gradio-container {
-        background-attachment: fixed !important;
-    }
+    background-attachment: fixed !important;
 }
 
 .gradio-container {
     width: 100% !important;
     max-width: 1200px !important;
-    min-width: 0 !important;
     margin: 0 auto !important;
     padding-top: 0 !important;
     padding-left: 20px !important;
     padding-right: 20px !important;
     box-sizing: border-box !important;
-    overflow-x: hidden;
+    overflow-x: clip;
 }
 .gradio-container .main,
 .gradio-container .wrap,
@@ -190,7 +166,6 @@ body, .gradio-container {
 .workspace-filters,
 .view-filters {
     max-width: 100% !important;
-    min-width: 0 !important;
 }
 
 /* —— App header (P-Bench only) —— */
@@ -325,9 +300,6 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
     background: transparent !important;
     box-shadow: none !important;
 }
-/* Flatten tabs so the bar sits above shared filters. display:contents is
-   the fallback; Safari can drop or mis-order those children, so browsers
-   with subgrid use the grid layout below instead. */
 .workspace-shell > .tabs,
 .workspace-shell > .main-tabs,
 .workspace-shell > .block:not(.workspace-filters),
@@ -351,49 +323,12 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
     margin: 0 0 16px !important;
     justify-content: center !important;
     width: 100% !important;
-    overflow: visible !important;
-}
-@supports (grid-template-rows: subgrid) {
-    .workspace-shell,
-    .workspace-shell.block,
-    .workspace-shell.column,
-    .workspace-shell.gap {
-        display: grid !important;
-        grid-template-columns: minmax(0, 1fr) !important;
-        grid-template-rows: auto auto auto !important;
-        align-content: start !important;
-    }
-    .workspace-shell > .tabs,
-    .workspace-shell > .main-tabs,
-    .workspace-shell .tabs.main-tabs {
-        display: grid !important;
-        grid-template-columns: minmax(0, 1fr) !important;
-        grid-template-rows: subgrid !important;
-        grid-column: 1 !important;
-        grid-row: 1 / 4 !important;
-        position: static !important;
-    }
-    .main-tabs > .tab-wrapper {
-        grid-row: 1 !important;
-        order: 0 !important;
-    }
-    .workspace-filters {
-        grid-column: 1 !important;
-        grid-row: 2 !important;
-        order: 0 !important;
-    }
-    .main-tabs .tabitem {
-        grid-row: 3 !important;
-        order: 0 !important;
-        min-width: 0 !important;
-    }
 }
 .main-tabs .tab-container {
     height: auto !important;
-    min-height: 0 !important;
     justify-content: center !important;
     flex-wrap: wrap !important;
-    overflow: visible !important;
+    overflow: hidden !important;
     max-width: 100% !important;
     gap: 2px;
 }
@@ -510,8 +445,6 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
     .app-header-brand h1,
     .gradio-container .app-header-brand h1 {
         font-size: 1.55rem !important;
-        width: auto !important;
-        max-width: 100% !important;
     }
     .app-header-tagline {
         font-size: 0.88rem !important;
@@ -588,29 +521,17 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
         left: 0 !important;
         width: 2.4rem;
         min-width: 2.4rem;
-        box-shadow: 6px 0 8px -6px rgba(0, 0, 0, 0.45);
     }
     .ranking-table .model-cell,
-    .prose .ranking-table .model-cell {
-        position: static !important;
-        left: auto !important;
-        z-index: auto;
-        min-width: 140px;
-        max-width: none;
-        background: transparent !important;
-    }
-    .ranking-table th.model-cell,
-    .prose .ranking-table th.model-cell {
+    .prose .ranking-table .model-cell,
+    .ranking-table th.model-cell {
         position: sticky !important;
-        top: 0 !important;
-        left: auto !important;
-        z-index: 3;
-        min-width: 140px;
-        max-width: none;
-        background: var(--pruna-bg-header) !important;
+        left: 2.4rem !important;
+        min-width: 108px;
+        max-width: 36vw;
     }
-    .ranking-table tbody tr:hover .model-cell {
-        background: var(--pruna-table-hover) !important;
+    .ranking-table .model-cell strong {
+        white-space: nowrap;
     }
     .compare-controls,
     .compare-controls.row,
@@ -683,7 +604,7 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
 .view-filters {
     display: flex !important;
     flex-wrap: wrap !important;
-    align-items: flex-end !important;
+    align-items: end !important;
     gap: 12px !important;
     margin: 0;
     overflow: visible !important;
@@ -691,7 +612,7 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
 .view-filters > div,
 .view-filters > .block,
 .view-filters > .form {
-    flex: 1 1 0% !important;
+    flex: 1 1 0 !important;
     min-width: 0 !important;
 }
 .view-filters > .block,
@@ -843,10 +764,6 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
     line-height: 1.45 !important;
     font-weight: 400 !important;
 }
-.view-help + .view-help,
-.prose .view-help + .view-help {
-    margin-top: 0.45rem !important;
-}
 .view-filters span[data-testid="block-info"],
 .view-filters .info,
 .view-filters .block-info {
@@ -984,7 +901,7 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
 .leaderboard-controls {
     display: flex !important;
     flex-wrap: wrap !important;
-    align-items: flex-end !important;
+    align-items: end !important;
     gap: 10px !important;
     margin-bottom: 12px;
     overflow: visible !important;
@@ -1400,9 +1317,7 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
     box-shadow: none !important;
 }
 .compare-controls .compare-prompt-count .head {
-    display: block !important;
-    grid-column: 1 / -1;
-    grid-row: 1;
+    display: contents;
     margin: 0 !important;
 }
 .compare-controls .compare-prompt-count .head label {
@@ -1576,182 +1491,7 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
     margin: 0 !important;
 }
 .compare-row { display: grid; gap: 12px; min-width: 0; width: 100%; }
-.compare-prompt-text { overflow-wrap: anywhere; word-break: break-word; }
-.pareto-heading-row,
-.pareto-heading-row.row,
-.pareto-heading-row .form {
-    display: flex !important;
-    flex-wrap: wrap !important;
-    align-items: center !important;
-    gap: 8px 12px !important;
-    width: 100% !important;
-    margin-bottom: 0.4rem !important;
-}
-.pareto-heading-row .pareto-subhead,
-.pareto-heading-row > div:first-child,
-.pareto-heading-row .form > div:first-child {
-    flex: 1 1 240px !important;
-    min-width: 0 !important;
-    margin: 0 !important;
-}
-.pareto-heading-row .pareto-scale-control {
-    display: flex !important;
-    flex-direction: row !important;
-    align-items: center !important;
-    justify-content: flex-end !important;
-    flex: 0 0 auto !important;
-    gap: 0 !important;
-    margin-left: auto !important;
-    max-width: 168px !important;
-    padding: 0 !important;
-}
-.pareto-scale-all-row,
-.pareto-scale-all-row.row,
-.pareto-scale-all-row .form {
-    display: flex !important;
-    flex-direction: row !important;
-    flex-wrap: wrap !important;
-    align-items: center !important;
-    justify-content: flex-start !important;
-    gap: 8px 12px !important;
-    width: 100% !important;
-    margin: 2px 0 14px !important;
-}
-.pareto-scale-all-row .html-container,
-.pareto-scale-all-row .block {
-    border: none !important;
-    background: transparent !important;
-    box-shadow: none !important;
-    padding: 0 !important;
-    margin: 0 !important;
-    width: auto !important;
-    flex: 0 0 auto !important;
-}
-.pareto-scale-all-row .html-container {
-    flex: 1 1 auto !important;
-    min-width: 0 !important;
-}
-.pareto-scale-all-label {
-    color: var(--pruna-text-muted);
-    font-size: 0.8rem;
-    font-weight: 500;
-    white-space: nowrap;
-}
-.pareto-scale-all-row .pareto-scale-toggle {
-    margin-left: auto !important;
-}
-.pareto-scale-toggle,
-.pareto-scale-toggle.block {
-    min-width: 0 !important;
-    width: auto !important;
-    border: none !important;
-    background: transparent !important;
-    box-shadow: none !important;
-    padding: 0 !important;
-    margin: 0 !important;
-}
-.pareto-scale-toggle .wrap,
-.pareto-scale-toggle .form {
-    display: block !important;
-    width: auto !important;
-    margin: 0 !important;
-    padding: 0 !important;
-    border: none !important;
-    background: transparent !important;
-    box-shadow: none !important;
-}
-.pareto-scale-toggle legend {
-    display: none !important;
-}
-.pareto-scale-toggle fieldset,
-.pareto-scale-toggle .wrap:has(> label),
-.pareto-scale-toggle .form:has(> label) {
-    position: relative !important;
-    display: grid !important;
-    grid-template-columns: 1fr 1fr !important;
-    align-items: stretch !important;
-    isolation: isolate;
-    box-sizing: border-box !important;
-    width: max-content !important;
-    min-width: 0 !important;
-    padding: 4px !important;
-    gap: 4px !important;
-    border: 1px solid var(--pruna-input-border) !important;
-    border-radius: 10px !important;
-    background: var(--pruna-toggle-track) !important;
-    box-shadow: none !important;
-}
-.pareto-scale-toggle fieldset::before,
-.pareto-scale-toggle .wrap:has(> label)::before,
-.pareto-scale-toggle .form:has(> label)::before {
-    content: none !important;
-}
-.pareto-scale-toggle label {
-    position: relative !important;
-    z-index: 1 !important;
-    display: flex !important;
-    flex: 1 1 auto !important;
-    align-items: center !important;
-    justify-content: center !important;
-    gap: 0 !important;
-    box-sizing: border-box !important;
-    min-width: 58px !important;
-    min-height: 26px !important;
-    margin: 0 !important;
-    padding: 5px 12px !important;
-    border: none !important;
-    border-radius: 6px !important;
-    background: transparent !important;
-    box-shadow: none !important;
-    color: var(--pruna-text-body) !important;
-    font-size: 0.72rem !important;
-    font-weight: 600 !important;
-    line-height: 1.2 !important;
-    letter-spacing: 0.01em;
-    white-space: nowrap;
-    cursor: pointer !important;
-}
-.pareto-scale-toggle-all label {
-    min-width: 68px !important;
-    min-height: 30px !important;
-    padding: 6px 14px !important;
-    font-size: 0.85rem !important;
-}
-.pareto-scale-toggle label span {
-    margin: 0 !important;
-    padding: 0 !important;
-    color: inherit !important;
-    opacity: 1 !important;
-}
-.pareto-scale-toggle label > * + * {
-    margin-left: 0 !important;
-}
-.pareto-scale-toggle label + label::before,
-.pareto-scale-toggle label + label {
-    content: none !important;
-    border-left: none !important;
-}
-.pareto-scale-toggle input[type="radio"] {
-    position: absolute !important;
-    appearance: none !important;
-    opacity: 0 !important;
-    width: 0 !important;
-    height: 0 !important;
-    margin: 0 !important;
-    pointer-events: none !important;
-}
-.pareto-scale-toggle label:hover {
-    background: transparent !important;
-    color: var(--pruna-text-primary) !important;
-}
-.pareto-scale-toggle label.selected,
-.pareto-scale-toggle label:has(input:checked) {
-    background: var(--pruna-toggle-thumb) !important;
-    color: var(--pruna-lavender) !important;
-    font-weight: 700 !important;
-    border-color: transparent !important;
-    box-shadow: var(--pruna-toggle-thumb-shadow) !important;
-}
+.compare-prompt-text { overflow-wrap: anywhere; }
 .pareto-layout,
 .pareto-layout.row,
 .pareto-layout .form {
@@ -1809,7 +1549,7 @@ button.theme-toggle[data-mode="light"] .theme-icon-moon { display: block !import
 .app-header .app-header-brand h1 {
     display: block !important;
     width: max-content !important;
-    max-width: 100% !important;
+    max-width: none !important;
     flex: 0 0 auto !important;
     margin: 0 !important;
     padding: 0 !important;
@@ -2286,7 +2026,7 @@ def load_qwen_combined_dataframe(path):
     df = df[~df["Model"].astype(str).str.startswith("#")].copy()
     df["Model"] = df["Model"].astype(str).str.strip()
 
-    df = _as_numeric(
+    return _as_numeric(
         df,
         [
             "Price / Image (USD)",
@@ -2296,9 +2036,7 @@ def load_qwen_combined_dataframe(path):
             "Rapidata Elo",
             "Datapoint Elo",
         ],
-    )
-    df = df.drop(columns=["Raw Win Rate"], errors="ignore")
-    return df.reset_index(drop=True)
+    ).reset_index(drop=True)
 
 
 df = load_oneig_dataframe(oneig_path)
@@ -2349,8 +2087,8 @@ qwen_combined_dir = _resolve_data_path(
     space_root.parent / "qwen_image_bench_combined",
 )
 qwen_path = _resolve_data_path(
-    data_dir / "qwen_image_bench_model_price_and_median_generation_time_10_august.csv",
-    space_root.parent / "qwen_image_bench_model_price_and_median_generation_time_10_august.csv",
+    data_dir / "qwen_image_bench_model_price_and_median_generation_time.csv",
+    space_root.parent / "qwen_image_bench_model_price_and_median_generation_time.csv",
 )
 aa_path = _resolve_data_path(
     data_dir / "artificial_analysis_text_to_image_leaderboard.csv",
@@ -2371,6 +2109,7 @@ qwen_display_columns = [
         "Datapoint Elo",
         "Rapidata Elo",
         "P-Judge Overall",
+        "Raw Win Rate",
         "Median Generation Time (s)",
         "Min Generation Time (s)",
         "Price / Image (USD)",
@@ -2578,23 +2317,11 @@ custom_head = """
     return found;
   };
 
-  const applyPlotTheme = (gd, mode) => {
-    const layout = PLOT_LAYOUT[mode];
-    if (!layout || typeof Plotly === "undefined" || !gd) return;
-    if (gd.layout && gd.layout.paper_bgcolor === layout.paper_bgcolor) return;
-    try { Plotly.relayout(gd, layout); } catch (e) {}
-  };
-
-  const watchPlotTheme = (gd) => {
-    if (!gd || gd.__inferbenchThemeBound) return;
-    gd.__inferbenchThemeBound = true;
-    gd.addEventListener("plotly_afterplot", () => applyPlotTheme(gd, currentMode()));
-  };
-
   const restylePlots = (mode) => {
+    const layout = PLOT_LAYOUT[mode];
+    if (!layout || typeof Plotly === "undefined") return;
     queryAll(".js-plotly-plot").forEach((gd) => {
-      watchPlotTheme(gd);
-      applyPlotTheme(gd, mode);
+      try { Plotly.relayout(gd, layout); } catch (e) {}
     });
   };
 

data/qwen_image_bench_model_price_and_median_generation_time.csv

@@ -0,0 +1,60 @@
+Model,Price / Image (USD),Median Generation Time (s),Min Generation Time (s),P-Judge Overall,Raw Win Rate,Rapidata Elo,Datapoint Elo,Benchmark.ai Elo
+reve_2_1,N/A,N/A,28.3,,,,,1173.2
+ideogram_4_0_quality,N/A,N/A,66.6,,,,,1131.0
+gpt_image_2,0.21,77.8,77.8,59.02083099999998,,1172.58,1116,1124.5
+nano_banana_2_0,N/A,N/A,N/A,56.362956,59.4%,1071.79,1067,1056.3
+gpt_image_1_5,0.135,38.0,38.0,57.864434999999986,64.5%,1102.09,1064,910.3
+hidream_i1_dev,0.0086,2.823569217998738,2.06,49.17682099999999,51.7%,999.17,984,
+gpt_image_1,0.167,38.8,38.8,54.975741000000006,,1095.39,987,
+imagen_4_0_ultra,0.06,11.4,11.4,53.385603999999965,59.7%,1075.22,1023,
+flux_2_flex,0.06,10.856533817990567,8.07,53.9245676767677,57.8%,1054.34,1018,921.3
+qwen_image,0.025,4.8,4.8,51.561746,51.5%,1073.72,1005,975.6
+seedream_5_0,N/A,N/A,N/A,55.715153,,1070.35,1012,
+nano_banana_pro,0.134,17.2,17.2,56.452189898989914,58.0%,1029.51,1045,1102.4
+hidream_i1_fast,0.0051,9.920469530501578,1.40,48.89298600000002,50.1%,1024.67,984,
+imagen_4_fast,0.02,3.7531301500021073,2.71,50.155055208333344,,972.6,981,
+seedream_4_5,0.04,16.6,16.6,55.66328800000001,,1048.96,1035,964.2
+seedream_4_0,0.03,12.2,12.2,55.241443999999994,,1050.41,1035,
+p_image_2_ideogram_low_1k,0.0075,2.59,1.55,54.827397,53.2%,1000.07,1009,1103.9
+#p_image_2_ideogram_low_2k,0.016,5.17,4.25,53.971723,48.4%,1074.91,1003,
+qwen_image_2_0_pro,0.035,35.5,35.5,56.181776,,1033.25,1017,959.9
+juggernaut_base_flux,0.035,4.115834823496698,3.84,49.474676,46.4%,1038.22,972,
+flux_2_pro,N/A,N/A,N/A,54.43423900000002,,993.54,1019,1011.6
+z_image,0.005,1.5122045120006078,1.22,49.946227999999984,48.1%,1028.1,1001,
+p_image_2_ideogram_high_1k,0.015,4.28,3.07,55.754507999999994,52.1%,972.92,1022,1104.0
+#p_image_2_ideogram_high_2k,0.03,8.69,7.12,54.757842000000004,49.4%,1074.91,1007,
+qwen_image_2512,0.02,19.1,19.1,51.677326,,1029.6,1009,
+flux_2_max,0.07,26.4,26.4,54.047976,60.9%,955.19,1028,1001.6
+flux_1_1_pro_ultra,0.06,9.026992494000297,6.20,50.358445,52.0%,979.56,995,
+juggernaut_pro_flux,0.055,3.699636150500737,3.36,49.741183,46.4%,972.14,964,
+flux_dev,0.025,1.6931055715031107,1.49,48.241479,42.1%,925.04,940,
+p_image_2_ideogram_very_low_1k,0.003,2.56,1.49,53.51894200000001,48.1%,959.86,995,1071.7
+#p_image_2_ideogram_very_low_2k,0.006,3.94,3.05,53.68248699999998,47.2%,962.38,980,
+#p_image_2_ideogram_very_low_1k_no_upsampling,0.005,0.82,,46.33,,,,1071.7
+#p_image_2_ideogram_very_low_2k_no_upsampling,0.005,2.18,,46.39,,,,1071.7
+#p_image_2_ideogram_low_1k_no_upsampling,0.01,3.33,,45.32,,,,1103.9
+#p_image_2_ideogram_low_2k_no_upsampling,0.01,3.34,,46.25,,,,1103.9
+#p_image_2_ideogram_medium_1k_no_upsampling,0.015,2.13,,46.88,,,,1115.1
+#p_image_2_ideogram_medium_2k_no_upsampling,0.015,5.55,,47.11,,,,1115.1
+#p_image_2_ideogram_high_1k_no_upsampling,0.03,3.11,,46.88,,,,1104.0
+#p_image_2_ideogram_high_2k_no_upsampling,0.03,5.55,,47.06,,,,1104.0
+hidream_i1_full,0.014,6.008430051002506,5.69,46.82559300000002,36.9%,955.46,944,
+flux_2_dev,0.025,4.310278721997747,4.03,52.71644489795918,53.1%,1007.6,1021,942.0
+imagen_4_0,0.04,14.1,14.1,52.08996199999999,53.5%,979.62,1005,
+wan_2_2_image,0.02,3.005390542501118,2.96,48.19959399999999,,944.87,960,
+flux_krea,0.025,1.7150160090022837,1.7150160090022837,50.35734,50.0%,919.73,975,
+p_image_2_ideogram_medium_1k,0.01,3.06,2.05,54.719193000000004,51.7%,941.46,1002,1115.1
+#p_image_2_ideogram_medium_2k,0.02,7.44,6.47,54.23124444444446,48.1%,949.35,1000,
+glm_image,0.05,188.2,188.2,51.42623399999999,,923.35,953,
+p_image,0.005,1.0640762715011078,0.95,48.75217099999999,44.8%,924.37,961,1098.7
+hunyuanimage_3_0,0.09,41.0,41.0,52.32440099999998,52.4%,1009.61,979,765.3
+juggernaut_lightning_flux,0.006,1.1787893719956628,0.93,48.30471699999998,40.5%,916.68,929,
+flux_1_1_pro,0.04,3.0104645500032348,2.34,49.92882700000001,50.6%,925.04,984,
+flux_schnell,0.003,0.8411653029907029,0.80,46.685981818181816,34.8%,892.32,915,
+kling_v2_1,N/A,N/A,N/A,51.044512,,870.04,981,
+#p_image_2_ideogram_very_high_high_1k,0.075,9.34,7.25,58.45,,,1025,
+#p_image_2_ideogram_very_high_low_1k,0.0375,26.17,10.44,57.59,,,1024,
+#p_image_2_ideogram_very_high_medium_1k,0.05,9.28,5.96,57.88,,,1020,
+#p_image_2_ideogram_very_high_very_low_1k,0.015,26.52,10.60,56.92,,,1002,
+#p_image_2_ideogram_final_1k,0.0375,11.17,5.55,58.28,,,,
+#p_image_2_ideogram_final_2k,0.075,14.34,9.96,57.68,,,,

data/qwen_image_bench_model_price_and_median_generation_time_10_august.csv

@@ -1,64 +0,0 @@
-Rapidata Model,Price / Image (USD),Median Generation Time (s),Min Generation Time (s),P-Judge Overall,Raw Win Rate,Rapidata Elo,Datapoint Elo,Benchmark.ai Elo
-reve_2_1,N/A,N/A,28.3,,,,,1173.2
-ideogram_4_0_quality,N/A,N/A,66.6,,,,,1131.0
-gpt_image_2,0.21,77.8,77.8,59.02083099999998,65.6%,1172.58,1110,1124.5
-nano_banana_2_0,N/A,N/A,N/A,56.362956,59.2%,1071.79,1063,1056.3
-gpt_image_1_5,0.135,38.0,38.0,57.864434999999986,58.9%,1102.09,1060,910.3
-hidream_i1_dev,0.0086,2.823569217998738,2.06,49.17682099999999,47.7%,999.17,983,
-gpt_image_1,0.167,38.8,38.8,54.975741000000006,47.4%,1095.39,984,
-imagen_4_0_ultra,0.06,11.4,11.4,53.385603999999965,52.9%,1075.22,1019,
-flux_2_flex,0.06,10.856533817990567,8.07,53.9245676767677,52.4%,1054.34,1014,921.3
-#flux_2_turbo,0.008,2.14,1.81,53.02,,,1003,
-#flux_2_flash,0.005,1.43,1.03,52.03,,,1001,
-qwen_image,0.025,4.8,4.8,51.561746,50.3%,1073.72,1002,975.6
-seedream_5_0,N/A,N/A,N/A,55.715153,51.6%,1070.35,1010,
-nano_banana_pro,0.134,17.2,17.2,56.452189898989914,56.4%,1029.51,1041,1102.4
-hidream_i1_fast,0.0051,9.920469530501578,1.40,48.89298600000002,47.2%,1024.67,980,
-imagen_4_fast,0.02,3.7531301500021073,2.71,50.155055208333344,47.2%,972.6,980,
-seedream_4_5,0.04,16.6,16.6,55.66328800000001,54.7%,1048.96,1032,964.2
-seedream_4_0,0.03,12.2,12.2,55.241443999999994,54.9%,1050.41,1032,
-p_image_2_ideogram_low_1k,0.0075,2.59,1.55,54.827397,50.8%,1000.07,1006,1103.9
-#p_image_2_ideogram_low_2k,0.016,5.17,4.25,53.971723,50.0%,1074.91,1000,1103.9
-qwen_image_2_0_pro,0.035,35.5,35.5,56.181776,52.2%,1033.25,1014,959.9
-juggernaut_base_flux,0.035,4.115834823496698,3.84,49.474676,46.1%,1038.22,971,
-flux_2_pro,N/A,N/A,N/A,54.43423900000002,52.2%,993.54,1014,1011.6
-z_image,0.005,1.5122045120006078,1.22,49.946227999999984,49.9%,1028.1,999,
-p_image_2_ideogram_high_1k,0.015,4.28,3.07,55.754507999999994,52.5%,972.92,1017,1104.0
-#p_image_2_ideogram_high_2k,0.06,8.69,7.12,54.757842000000004,50.7%,1074.91,1006,1104.0
-qwen_image_2512,0.02,19.1,19.1,51.677326,50.9%,1029.6,1005,
-flux_2_max,0.07,26.4,26.4,54.047976,53.9%,955.19,1026,1001.6
-flux_1_1_pro_ultra,0.06,9.026992494000297,6.20,50.358445,48.6%,979.56,990,
-juggernaut_pro_flux,0.055,3.699636150500737,3.36,49.741183,44.7%,972.14,963,
-flux_dev,0.025,1.6931055715031107,1.49,48.241479,41.5%,925.04,941,
-p_image_2_ideogram_very_low_1k,0.003,2.56,1.49,53.51894200000001,49.1%,959.86,994,1071.7
-#p_image_2_ideogram_very_low_2k,0.006,3.94,3.05,53.68248699999998,46.3%,962.38,976,1071.7
-#p_image_2_ideogram_very_low_1k_no_upsampling,0.005,0.82,,46.33,,,,1071.7
-#p_image_2_ideogram_very_low_2k_no_upsampling,0.005,2.18,,46.39,,,,1071.7
-#p_image_2_ideogram_low_1k_no_upsampling,0.01,3.33,,45.32,,,,1103.9
-#p_image_2_ideogram_low_2k_no_upsampling,0.01,3.34,,46.25,,,,1103.9
-#p_image_2_ideogram_medium_1k_no_upsampling,0.015,2.13,,46.88,,,,1115.1
-#p_image_2_ideogram_medium_2k_no_upsampling,0.015,5.55,,47.11,,,,1115.1
-#p_image_2_ideogram_high_1k_no_upsampling,0.03,3.11,,46.88,,,,1104.0
-#p_image_2_ideogram_high_2k_no_upsampling,0.03,5.55,,47.06,,,,1104.0
-hidream_i1_full,0.014,6.008430051002506,5.69,46.82559300000002,41.4%,955.46,941,
-flux_2_dev,0.025,4.310278721997747,4.03,52.71644489795918,52.5%,1007.6,1016,942.0
-imagen_4_0,0.04,14.1,14.1,52.08996199999999,50.4%,979.62,1002,
-wan_2_2_image,0.02,3.005390542501118,2.96,48.19959399999999,44.1%,944.87,958,
-flux_krea,0.025,1.7150160090022837,1.7150160090022837,50.35734,46.2%,919.73,973,
-p_image_2_ideogram_medium_1k,0.01,3.06,2.05,54.719193000000004,49.8%,941.46,999,1115.1
-#p_image_2_ideogram_medium_2k,0.02,7.44,6.47,54.23124444444446,49.3%,949.35,996,1115.1
-glm_image,0.05,188.2,188.2,51.42623399999999,42.9%,923.35,950,
-p_image,0.005,1.0640762715011078,0.95,48.75217099999999,44.3%,924.37,960,1098.7
-hunyuanimage_3_0,0.09,41.0,41.0,52.32440099999998,46.6%,1009.61,977,765.3
-juggernaut_lightning_flux,0.006,1.1787893719956628,0.93,48.30471699999998,40.0%,916.68,930,
-flux_1_1_pro,0.04,3.0104645500032348,2.34,49.92882700000001,47.7%,925.04,984,
-flux_schnell,0.003,0.8411653029907029,0.80,46.685981818181816,37.6%,892.32,914,
-kling_v2_1,N/A,N/A,N/A,51.044512,47.2%,870.04,980,
-#p_image_2_ideogram_very_high_high_1k,0.075,9.34,7.25,58.45,52.9%,,1024,
-#p_image_2_ideogram_very_high_low_1k,0.0375,26.17,10.44,57.59,52.9%,,1023,
-#p_image_2_ideogram_very_high_medium_1k,0.05,9.28,5.96,57.88,52.3%,,1019,
-#p_image_2_ideogram_very_high_very_low_1k,0.015,26.52,10.60,56.92,49.4%,,1001,
-p_image_2_ideogram_very_high_1k,0.033,11.17,5.55,58.28,55.5%,,1034,
-#p_image_2_ideogram_very_high_2k,0.066,14.34,9.96,57.68,53.4%,,1020,
-#p_image_2_ideogram_very_high_john_1k,,8.53,6.43,58.19,,,1015,
-#p_image_2_ideogram_very_high_john_2k,,13.86,10.64,57.75,,,1002,

model_display.py

@@ -87,7 +87,6 @@ MODEL_DISPLAY_NAMES = {
     "p_image_2_ideogram_high_1k": "P-Image-Ideogram High 1K",
     "p_image_2_ideogram_high_2k": "P-Image-Ideogram High 2K",
     "P-Image-Ideogram (High)": "P-Image-Ideogram High",
-    "p_image_2_ideogram_very_high_1k": "P-Image-Ideogram Very High 1K",
     # Others overlapping P-Bench
     "z_image": "Z-Image",
     "glm_image": "GLM-Image",

ui.py

@@ -1,5 +1,4 @@
 from html import escape
-from math import ceil, floor, log10
 from pathlib import Path
 import base64
 import random
@@ -26,12 +25,6 @@ MAX_PARETO_METRICS = 8
 _PARETO_SLOT_COUNT = 1 + MAX_PARETO_METRICS * 8
 _PARETO_PRICE_COLUMN = "Price / Image (USD)"
 _PARETO_TIME_COLUMN = "Min Generation Time (s)"
-_PARETO_SCALE_CHOICES = [
-    ("Log", "Logarithmic"),
-    ("Linear", "Linear"),
-]
-_PARETO_SCALE_VALUES = {value for _, value in _PARETO_SCALE_CHOICES}
-_PARETO_SCALE_DEFAULT = "Logarithmic"
 
 TAB_LEADERBOARDS = "leaderboards"
 TAB_PARETO = "pareto"
@@ -251,14 +244,6 @@ def _dataset_has_samples(datasets, dataset_id):
     return bool(dataset and dataset.get("samples"))
 
 
-def _sample_model_ids(datasets, dataset_id):
-    dataset = _item(datasets, dataset_id)
-    samples = dataset.get("samples") if dataset else None
-    if not samples:
-        return set()
-    return set(samples.get("models") or [])
-
-
 def _dataset_has_pareto(datasets, dataset_id):
     dataset = _item(datasets, dataset_id)
     columns = getattr(dataset.get("data") if dataset else None, "columns", [])
@@ -329,25 +314,23 @@ def _metric_dropdown_value(metric_id):
     ]
 
 
-def _model_choices(datasets, dataset_id, *, require_samples=False):
+def _model_choices(datasets, dataset_id):
     cached = _MODEL_CHOICES_CACHE.get(dataset_id)
-    if cached is None:
-        dataset = _item(datasets, dataset_id)
-        data = dataset.get("data") if dataset else None
-        if data is None or "Model" not in getattr(data, "columns", []):
-            cached = []
-        else:
-            models = data["Model"].dropna().astype(str).unique().tolist()
-            # (label, value) so the UI shows the shared name but filters on the raw id.
-            cached = sorted(
-                ((display_model_name(model), model) for model in models),
-                key=lambda item: item[0].casefold(),
-            )
-        _MODEL_CHOICES_CACHE[dataset_id] = cached
-    if not require_samples:
+    if cached is not None:
         return cached
-    allowed = _sample_model_ids(datasets, dataset_id)
-    return [choice for choice in cached if choice[1] in allowed]
+    dataset = _item(datasets, dataset_id)
+    data = dataset.get("data") if dataset else None
+    if data is None or "Model" not in getattr(data, "columns", []):
+        _MODEL_CHOICES_CACHE[dataset_id] = []
+        return []
+    models = data["Model"].dropna().astype(str).unique().tolist()
+    # (label, value) so the UI shows the shared name but filters on the raw id.
+    choices = sorted(
+        ((display_model_name(model), model) for model in models),
+        key=lambda item: item[0].casefold(),
+    )
+    _MODEL_CHOICES_CACHE[dataset_id] = choices
+    return choices
 
 
 def _model_choice_values(choices):
@@ -602,6 +585,7 @@ def _display_label(column):
         "Arena Art Elo": "Art",
         "Arena Portraits Elo": "Portraits",
         "Arena Text Rendering Elo": "Text Rendering",
+        "Raw Win Rate": "Raw win rate",
         "Median Generation Time (s)": "Median generation time",
         "Min Generation Time (s)": "Min generation time",
         "Price / Image (USD)": "Price per image",
@@ -686,7 +670,6 @@ def _build_pareto_figure(
     x_title,
     x_hover_prefix="",
     x_hover_suffix="",
-    x_axis_type="linear",
 ):
     scatter = (
         data[["Model", score_column, x_column]]
@@ -772,29 +755,7 @@ def _build_pareto_figure(
     )
     axis_font = {"color": "#fafafa", "size": 13}
     tick_font = {"color": "#a3a3a3", "size": 12}
-    x_axis_ticks = {}
-    if x_axis_type == "log":
-        positive_x = scatter.loc[scatter[x_column] > 0, x_column].astype(float)
-        if not positive_x.empty:
-            minimum = positive_x.min()
-            maximum = positive_x.max()
-            tick_values = [
-                factor * (10**exponent)
-                for exponent in range(
-                    floor(log10(minimum)),
-                    ceil(log10(maximum)) + 1,
-                )
-                for factor in (1, 2, 5)
-                if minimum * 0.8 <= factor * (10**exponent) <= maximum * 1.2
-            ]
-            x_axis_ticks = {
-                "tickmode": "array",
-                "tickvals": tick_values,
-                "ticktext": [f"{value:g}" for value in tick_values],
-            }
     fig.update_xaxes(
-        type=x_axis_type,
-        **x_axis_ticks,
         showgrid=True,
         gridcolor="rgba(74, 57, 98, 0.55)",
         zeroline=False,
@@ -813,55 +774,6 @@ def _build_pareto_figure(
     return fig
 
 
-def _is_log_scale(scale):
-    return scale == "Logarithmic"
-
-
-def _pareto_axis_type(scale):
-    return "log" if _is_log_scale(scale) else "linear"
-
-
-def _pareto_scale_radio(*extra_classes):
-    return gr.Radio(
-        choices=_PARETO_SCALE_CHOICES,
-        value=_PARETO_SCALE_DEFAULT,
-        show_label=False,
-        container=False,
-        elem_classes=["pareto-scale-toggle", *extra_classes],
-    )
-
-
-def _pareto_plot_heading(title):
-    with gr.Row(equal_height=False, elem_classes="pareto-heading-row"):
-        gr.Markdown(f"#### {title}", elem_classes="pareto-subhead")
-        with gr.Column(min_width=140, elem_classes="pareto-scale-control"):
-            return _pareto_scale_radio()
-
-
-def _default_pareto_scales():
-    return [_PARETO_SCALE_DEFAULT] * MAX_PARETO_METRICS
-
-
-def _normalize_pareto_scales(scales):
-    values = list(scales or [])
-    if len(values) < MAX_PARETO_METRICS:
-        values.extend(
-            [_PARETO_SCALE_DEFAULT] * (MAX_PARETO_METRICS - len(values))
-        )
-    return values[:MAX_PARETO_METRICS]
-
-
-def _uniform_pareto_scales(scale):
-    return [scale] * MAX_PARETO_METRICS
-
-
-def _pareto_master_scale_update(price_scales, time_scales):
-    values = list(price_scales) + list(time_scales)
-    if values and all(value == values[0] for value in values):
-        return gr.update(value=values[0])
-    return gr.update(value=None)
-
-
 def _pareto_axis(data, score_column, x_column, x_title, missing_message, empty_message, **hover):
     if x_column not in data.columns:
         return None, missing_message
@@ -877,12 +789,7 @@ def _pareto_axis(data, score_column, x_column, x_title, missing_message, empty_m
     return fig, None
 
 
-def _pareto_pair(
-    data,
-    score_column,
-    latency_scale=_PARETO_SCALE_DEFAULT,
-    price_scale=_PARETO_SCALE_DEFAULT,
-):
+def _pareto_pair(data, score_column):
     score_missing = "No score data is available for this metric."
     if data is None or not score_column or score_column not in data.columns:
         return None, score_missing, None, score_missing
@@ -895,7 +802,6 @@ def _pareto_pair(
         "Price per image isn't available for this dataset.",
         "No models have both a score and a price for this metric.",
         x_hover_prefix="$",
-        x_axis_type=_pareto_axis_type(price_scale),
     )
     time_fig, time_message = _pareto_axis(
         data,
@@ -905,7 +811,6 @@ def _pareto_pair(
         "Min generation time isn't available for this dataset.",
         "No models have both a score and a min generation time for this metric.",
         x_hover_suffix="s",
-        x_axis_type=_pareto_axis_type(latency_scale),
     )
     return price_fig, price_message, time_fig, time_message
 
@@ -944,16 +849,9 @@ def _pareto_slot_note(price_fig, price_message, time_fig, time_message, data):
     return " ".join(notes)
 
 
-def _pareto_slot_updates(
-    data,
-    score_columns,
-    price_scales=None,
-    time_scales=None,
-):
+def _pareto_slot_updates(data, score_columns):
     """Updates for a fixed bank of Gradio Plot slots (visible/hidden)."""
     score_columns = [column for column in (score_columns or []) if column]
-    price_scales = _normalize_pareto_scales(price_scales)
-    time_scales = _normalize_pareto_scales(time_scales)
     has_price = data is not None and _PARETO_PRICE_COLUMN in data.columns
     has_time = data is not None and _PARETO_TIME_COLUMN in data.columns
     dataset_note = _pareto_dataset_message(data)
@@ -975,10 +873,7 @@ def _pareto_slot_updates(
             continue
         score_column = score_columns[index]
         price_fig, price_message, time_fig, time_message = _pareto_pair(
-            data,
-            score_column,
-            latency_scale=time_scales[index],
-            price_scale=price_scales[index],
+            data, score_column
         )
         show_price = price_fig is not None
         show_time = time_fig is not None
@@ -1005,30 +900,6 @@ def _pareto_slot_updates(
     return updates
 
 
-def _pareto_all_scale_updates(data, score_columns, scale):
-    """Apply one scale to every Pareto plot and radio."""
-    score_columns = [column for column in (score_columns or []) if column]
-    price_updates = []
-    time_updates = []
-    for index in range(MAX_PARETO_METRICS):
-        if index >= len(score_columns):
-            price_updates.append(gr.skip())
-            time_updates.append(gr.skip())
-            continue
-        price_fig, _, time_fig, _ = _pareto_pair(
-            data,
-            score_columns[index],
-            latency_scale=scale,
-            price_scale=scale,
-        )
-        price_updates.append(_pareto_plot_update(price_fig))
-        time_updates.append(_pareto_plot_update(time_fig))
-    radio_updates = [
-        gr.update(value=scale) for _ in range(MAX_PARETO_METRICS * 2)
-    ]
-    return price_updates + time_updates + radio_updates
-
-
 def _samples_html(samples, selected_models, num_prompts, seed=0):
     if not samples:
         return _pareto_unavailable_html(
@@ -1173,10 +1044,10 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             gr.Markdown(
                 "<p class='filter-help'>"
                 "These filters apply to Leaderboards, Pareto plots, and Samples. "
-                "On Samples, only datasets and models we have generations for "
-                "are listed. On Pareto plots, only datasets with price or "
-                "generation time are listed. Search in Models, or leave it "
-                "empty to include every model."
+                "On Samples, only datasets we have generations for are listed. "
+                "On Pareto plots, only datasets with price or generation time "
+                "are listed. Search in Models, or leave it empty to include "
+                "every model."
                 "</p>",
                 elem_classes="filter-help-host",
             )
@@ -1246,27 +1117,12 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             ) as pp_tab:
                 gr.Markdown(
                     "<p class='view-help'>"
-                    "Score against price and generation time. Green points are on "
-                    "the frontier; lavender points sit below it. Hover a point to "
-                    "see which model it is."
-                    "</p>"
-                    "<p class='view-help'>"
-                    "Plots use a logarithmic scale by default. You can switch "
-                    "to linear for all plots, or individually for each plot."
+                    "Score against price and generation time. Green points are on the "
+                    "frontier; lavender points sit below it. Hover a point to see "
+                    "which model it is."
                     "</p>",
                     elem_classes="view-help-host",
                 )
-                with gr.Row(
-                    equal_height=False,
-                    elem_classes="pareto-scale-all-row",
-                ):
-                    gr.HTML(
-                        "<span class='pareto-scale-all-label'>All plots</span>",
-                        padding=False,
-                    )
-                    pareto_all_scale = _pareto_scale_radio(
-                        "pareto-scale-toggle-all",
-                    )
                 pareto_dataset_note = gr.HTML(
                     "",
                     padding=False,
@@ -1292,8 +1148,9 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
                                 min_width=320,
                                 elem_classes="pareto-col",
                             ) as slot_price_col:
-                                slot_price_scale = _pareto_plot_heading(
-                                    "Price vs score"
+                                gr.Markdown(
+                                    "#### Price vs score",
+                                    elem_classes="pareto-subhead",
                                 )
                                 slot_price = gr.Plot(
                                     value=None,
@@ -1305,8 +1162,9 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
                                 min_width=320,
                                 elem_classes="pareto-col",
                             ) as slot_time_col:
-                                slot_time_scale = _pareto_plot_heading(
-                                    "Min generation time vs score"
+                                gr.Markdown(
+                                    "#### Min generation time vs score",
+                                    elem_classes="pareto-subhead",
                                 )
                                 slot_time = gr.Plot(
                                     value=None,
@@ -1327,10 +1185,8 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
                             slot_layout,
                             slot_price_col,
                             slot_price,
-                            slot_price_scale,
                             slot_time_col,
                             slot_time,
-                            slot_time_scale,
                         )
                     )
 
@@ -1379,16 +1235,12 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             with gr.TabItem("About", id=TAB_ABOUT) as about_tab:
                 render_about()
 
-    def _synced_filters(
-        dataset_id, metric_id, models, *, clear_metric=False, require_samples=False
-    ):
+    def _synced_filters(dataset_id, metric_id, models, *, clear_metric=False):
         if clear_metric:
             metric_id = []
         else:
             metric_id = _coerce_metric(datasets, metrics, dataset_id, metric_id)
-        model_choices = _model_choices(
-            datasets, dataset_id, require_samples=require_samples
-        )
+        model_choices = _model_choices(datasets, dataset_id)
         model_values = set(_model_choice_values(model_choices))
         models = [model for model in (models or []) if model in model_values]
         metric_choices = _metric_dropdown_choices(datasets, metrics, dataset_id)
@@ -1472,8 +1324,6 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             "optimized": list(
                 extras.get("optimized", prev.get("optimized") or [])
             ),
-            "price_scales": _normalize_pareto_scales(prev.get("price_scales")),
-            "time_scales": _normalize_pareto_scales(prev.get("time_scales")),
             "stale": {
                 TAB_LEADERBOARDS: not flags["include_leaderboard"],
                 TAB_PARETO: not flags["include_pareto"],
@@ -1533,8 +1383,6 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
         include_leaderboard=True,
         include_pareto=False,
         include_samples=False,
-        price_scales=None,
-        time_scales=None,
     ):
         view = resolve_view(datasets, metrics, dataset_id, metric_id)
         data = view["data"]
@@ -1555,12 +1403,7 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             ranking_html = gr.skip()
         if include_pareto:
             pareto_data = _filter_leaderboard(data, [], [], [], models=models)
-            pareto_updates = _pareto_slot_updates(
-                pareto_data,
-                view["score_columns"],
-                price_scales=price_scales,
-                time_scales=time_scales,
-            )
+            pareto_updates = _pareto_slot_updates(pareto_data, view["score_columns"])
         else:
             pareto_updates = _pareto_skip_updates()
         if include_samples:
@@ -1602,20 +1445,10 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
         dataset_changed = source == "dataset" and dataset_id != view_state.get(
             "dataset_id"
         )
-        can_pareto = _dataset_has_pareto(datasets, dataset_id)
-        can_samples = _dataset_has_samples(datasets, dataset_id)
-        selected_tab = tab
+
         if source == "dataset":
-            if tab == TAB_SAMPLES and not can_samples:
-                selected_tab = TAB_LEADERBOARDS
-            elif tab == TAB_PARETO and not can_pareto:
-                selected_tab = TAB_LEADERBOARDS
             synced = _synced_filters(
-                dataset_id,
-                metric_id,
-                models,
-                clear_metric=dataset_changed,
-                require_samples=selected_tab == TAB_SAMPLES,
+                dataset_id, metric_id, models, clear_metric=dataset_changed
             )
             dataset_id, metric_id, models = synced[:3]
             metric_update, models_update = synced[3], synced[4]
@@ -1653,13 +1486,20 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
         ):
             return None
 
+        selected_tab = tab
         extras = (
             list(platform_value or []),
             list(owner_value or []),
             list(optimized_value or []),
         )
         extra_updates = None
+        can_pareto = _dataset_has_pareto(datasets, dataset_id)
+        can_samples = _dataset_has_samples(datasets, dataset_id)
         if source == "dataset":
+            if tab == TAB_SAMPLES and not can_samples:
+                selected_tab = TAB_LEADERBOARDS
+            elif tab == TAB_PARETO and not can_pareto:
+                selected_tab = TAB_LEADERBOARDS
             view = resolve_view(datasets, metrics, dataset_id, metric_id)
             extra_updates = _leaderboard_extras(
                 view["data"] if view else None,
@@ -1697,8 +1537,6 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
                 extras[2],
                 num_prompts,
                 seed,
-                price_scales=view_state.get("price_scales"),
-                time_scales=view_state.get("time_scales"),
                 **flags,
             ),
             "state": _commit_state(
@@ -1839,17 +1677,7 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             )
         dataset_update = _dataset_dropdown_update(datasets, tab, dataset_id)
         metric_id = _coerce_metric(datasets, metrics, dataset_id, metric_id)
-        require_samples = tab == TAB_SAMPLES
-        model_choices = _model_choices(
-            datasets, dataset_id, require_samples=require_samples
-        )
-        allowed_models = set(_model_choice_values(model_choices))
-        models = [model for model in (models or []) if model in allowed_models]
-        models_update = (
-            gr.update(choices=model_choices, value=models)
-            if (prev_tab == TAB_SAMPLES) != require_samples
-            else gr.skip()
-        )
+        models = list(models or [])
         view_state["current_tab"] = tab
         view_state["dataset_id"] = dataset_id
         view_state["metric_id"] = metric_id
@@ -1886,7 +1714,7 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             filters_vis,
             dataset_update,
             metric_vis,
-            models_update,
+            gr.skip(),
             *lb_filters,
         )
         tab_select = (
@@ -1911,8 +1739,6 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             optimized_value,
             num_prompts,
             seed,
-            price_scales=view_state.get("price_scales"),
-            time_scales=view_state.get("time_scales"),
             **flags,
         )
         stale[tab] = False
@@ -1969,70 +1795,6 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             next_seed,
         )
 
-    def _on_pareto_plot_scale(slot_index, axis):
-        def handler(dataset_id, metric_id, models, scale, view_state):
-            view_state = dict(view_state or {})
-            price_scales = _normalize_pareto_scales(
-                view_state.get("price_scales")
-            )
-            time_scales = _normalize_pareto_scales(
-                view_state.get("time_scales")
-            )
-            if axis == "price":
-                if price_scales[slot_index] == scale:
-                    return gr.skip(), gr.skip(), gr.skip()
-                price_scales[slot_index] = scale
-            else:
-                if time_scales[slot_index] == scale:
-                    return gr.skip(), gr.skip(), gr.skip()
-                time_scales[slot_index] = scale
-            view_state["price_scales"] = price_scales
-            view_state["time_scales"] = time_scales
-            master_scale = _pareto_master_scale_update(
-                price_scales, time_scales
-            )
-            view = resolve_view(datasets, metrics, dataset_id, metric_id)
-            score_columns = [
-                column for column in (view["score_columns"] or []) if column
-            ]
-            if slot_index >= len(score_columns):
-                return gr.skip(), master_scale, view_state
-            data = _filter_leaderboard(
-                view["data"], [], [], [], models=list(models or [])
-            )
-            price_fig, _, time_fig, _ = _pareto_pair(
-                data,
-                score_columns[slot_index],
-                latency_scale=time_scales[slot_index],
-                price_scale=price_scales[slot_index],
-            )
-            fig = price_fig if axis == "price" else time_fig
-            return _pareto_plot_update(fig), master_scale, view_state
-
-        handler.__name__ = f"on_pareto_{axis}_scale_{slot_index}"
-        return handler
-
-    def on_pareto_all_scale(dataset_id, metric_id, models, scale, view_state):
-        if scale not in _PARETO_SCALE_VALUES:
-            return (*_skip_all(MAX_PARETO_METRICS * 4), gr.skip())
-        view_state = dict(view_state or {})
-        scales = _uniform_pareto_scales(scale)
-        if (
-            _normalize_pareto_scales(view_state.get("price_scales")) == scales
-            and _normalize_pareto_scales(view_state.get("time_scales")) == scales
-        ):
-            return (*_skip_all(MAX_PARETO_METRICS * 4), gr.skip())
-        view_state["price_scales"] = scales
-        view_state["time_scales"] = scales
-        view = resolve_view(datasets, metrics, dataset_id, metric_id)
-        data = _filter_leaderboard(
-            view["data"], [], [], [], models=list(models or [])
-        )
-        return (
-            *_pareto_all_scale_updates(data, view["score_columns"], scale),
-            view_state,
-        )
-
     def _on_tab(tab):
         def handler(
             dataset_id,
@@ -2070,8 +1832,6 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             "platform": [],
             "owner": [],
             "optimized": [],
-            "price_scales": _default_pareto_scales(),
-            "time_scales": _default_pareto_scales(),
             "stale": {
                 TAB_LEADERBOARDS: False,
                 TAB_PARETO: True,
@@ -2083,7 +1843,7 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
         pareto_dataset_note,
         *[
             component
-            for slot_group, slot_title, slot_note, slot_layout, slot_price_col, slot_price, slot_price_scale, slot_time_col, slot_time, slot_time_scale in pareto_slots
+            for slot_group, slot_title, slot_note, slot_layout, slot_price_col, slot_price, slot_time_col, slot_time in pareto_slots
             for component in (
                 slot_group,
                 slot_title,
@@ -2096,24 +1856,6 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             )
         ],
     ]
-    pareto_all_scale_outputs = [
-        *[
-            slot_price
-            for _, _, _, _, _, slot_price, _, _, _, _ in pareto_slots
-        ],
-        *[
-            slot_time
-            for _, _, _, _, _, _, _, _, slot_time, _ in pareto_slots
-        ],
-        *[
-            slot_price_scale
-            for _, _, _, _, _, _, slot_price_scale, _, _, _ in pareto_slots
-        ],
-        *[
-            slot_time_scale
-            for _, _, _, _, _, _, _, _, _, slot_time_scale in pareto_slots
-        ],
-    ]
     view_inputs = [
         platform,
         owner,
@@ -2218,56 +1960,6 @@ def render_image_workspace(datasets, metrics, default_dataset_id, default_metric
             show_progress="hidden",
         )
 
-    pareto_all_scale.change(
-        on_pareto_all_scale,
-        inputs=[
-            dataset_dd,
-            metric_dd,
-            models_dd,
-            pareto_all_scale,
-            view_state,
-        ],
-        outputs=[*pareto_all_scale_outputs, view_state],
-        **_VIEW_EVENTS,
-    )
-
-    for slot_index, (
-        _,
-        _,
-        _,
-        _,
-        _,
-        slot_price,
-        slot_price_scale,
-        _,
-        slot_time,
-        slot_time_scale,
-    ) in enumerate(pareto_slots):
-        slot_price_scale.change(
-            _on_pareto_plot_scale(slot_index, "price"),
-            inputs=[
-                dataset_dd,
-                metric_dd,
-                models_dd,
-                slot_price_scale,
-                view_state,
-            ],
-            outputs=[slot_price, pareto_all_scale, view_state],
-            **_VIEW_EVENTS,
-        )
-        slot_time_scale.change(
-            _on_pareto_plot_scale(slot_index, "time"),
-            inputs=[
-                dataset_dd,
-                metric_dd,
-                models_dd,
-                slot_time_scale,
-                view_state,
-            ],
-            outputs=[slot_time, pareto_all_scale, view_state],
-            **_VIEW_EVENTS,
-        )
-
     prompt_count.change(
         on_samples_controls,
         inputs=[dataset_dd, models_dd, prompt_count, seed_state],