Fix model catalog server auto-selection

CHANGELOG.md

@@ -16,6 +16,7 @@ The format is based on Keep a Changelog and this project follows Semantic Versio
 - **Onboarding prompt scope** — the Run page completion handoff now appears only for the onboarding first-run step, canceling the onboarding-launched server drawer stops setup with an explicit normal-mode notice, and the three-step welcome layout is centered.
 - **Built-in template reload after DB clear** — clearing the database from Settings now reloads the built-in benchmark library immediately, keeping shipped templates available without restarting the backend.
 - **Catalog empty server card** — the empty Servers catalog now presents a dashed first-server card with the add action instead of a centered empty-state panel.
+- **Catalog model auto-selection** — opening the Models catalog without a server filter now selects the first available inference server so discovered models render immediately.
 - **Run empty preview** — the empty Run workspace now shows a dummy benchmark result with sample model, prompt, metrics, and audit rows instead of a generic empty panel.
 
 ## [0.9.0] - 2026-06-17

frontend/src/pages/Catalog.tsx

@@ -613,6 +613,12 @@ export function Catalog({
   const selectedQuantizations = useMemo(() => new Set(parseCsv(searchParams.get('quantization'))), [searchParams]);
   const selectedFormats = useMemo(() => new Set(parseCsv(searchParams.get('format'))), [searchParams]);
   const selectedCapabilities = useMemo(() => new Set(parseCsv(searchParams.get('capabilities'))), [searchParams]);
+  const autoSelectedModelServer = useMemo(() => {
+    return servers.find((server) => server.inference_server.active && !server.inference_server.archived)
+      ?? servers.find((server) => !server.inference_server.archived)
+      ?? servers[0]
+      ?? null;
+  }, [servers]);
   const maxParamCount = useMemo(() => {
     const v = searchParams.get('maxParams');
     return v !== null ? Number(v) : null;
@@ -626,6 +632,16 @@ export function Catalog({
     }, { replace: true });
   }, [activeTab, inspectorModelId, inspectorServerId, navigate]);
 
+  useEffect(() => {
+    if (activeTab !== 'models' || inspectorServerId || inspectorModelId || searchParams.has('servers') || !autoSelectedModelServer) {
+      return;
+    }
+    const next = new URLSearchParams(searchParams);
+    next.set('tab', 'models');
+    next.set('servers', autoSelectedModelServer.inference_server.server_id);
+    setSearchParams(next, { replace: true });
+  }, [activeTab, autoSelectedModelServer, inspectorModelId, inspectorServerId, searchParams, setSearchParams]);
+
   useEffect(() => {
     if (searchParams.get('startOnboarding') !== '1') {
       return;
@@ -903,7 +919,7 @@ export function Catalog({
           setServerStageCollapsed={setServerStageCollapsed}
           onToggleServer={toggleServerSelection}
           onClearServers={() => updateQuery((params) => {
-            params.delete('servers');
+            params.set('servers', '');
             params.delete('family');
             params.delete('quantization');
             params.delete('format');

frontend/tests/e2e/models.spec.ts

@@ -64,19 +64,18 @@ async function seedModel(
   }
 }
 
-test('models tab requires a selected server before showing model cards', async ({ page, request }) => {
-  const server = await createInferenceServer(request);
+test('models tab auto-selects an available server before showing model cards', async ({ page, request }) => {
+  const server = await createInferenceServer(request, { display_name: `! Auto-select ${Date.now()}` });
   await seedModel(request, server.inference_server.server_id, {
     model_id: '/lmstudio-community/Qwen3-Coder-30B-A3B-Instruct-MLX-6bit',
     display_name: 'Qwen3 Coder',
     format: 'MLX'
   });
 
   await page.goto('/catalog?tab=models');
-  await expect(page.getByRole('heading', { name: 'Select a server to see its models' })).toBeVisible();
-
-  await page.locator('.server-filter-row').filter({ hasText: server.inference_server.display_name }).click();
   await expect(page.locator('.catalog-model-card').filter({ hasText: 'Qwen3 Coder' })).toBeVisible();
+  await expect(page.locator('.server-filter-row.is-selected').filter({ hasText: server.inference_server.display_name })).toBeVisible();
+  await expect(page).toHaveURL(new RegExp(`servers=${encodeURIComponent(server.inference_server.server_id)}`));
 
   await archiveInferenceServer(request, server.inference_server.server_id);
 });