diff --git a/src/api/services/index.ts b/src/api/services/index.ts
index fb6ad360..8b213e17 100644
--- a/src/api/services/index.ts
+++ b/src/api/services/index.ts
@@ -64,6 +64,12 @@ export {
 // OpenRouter catalog and picker
 export { isOpenRouterUrl, fetchOpenRouterModels, type OpenRouterModel } from './openrouter-catalog';
 export { pickOpenRouterModel, type OpenRouterSelection } from './openrouter-picker';
+export {
+  getLocalRuntimeReadiness,
+  type LocalRuntimeId,
+  type LocalRuntimeReadiness,
+  type LocalRuntimeStatus,
+} from './local-runtime-readiness';
 
 // Provider presets for CLI
 export {
diff --git a/src/api/services/local-runtime-readiness.ts b/src/api/services/local-runtime-readiness.ts
new file mode 100644
index 00000000..0472bfc1
--- /dev/null
+++ b/src/api/services/local-runtime-readiness.ts
@@ -0,0 +1,121 @@
+export type LocalRuntimeId = 'ollama' | 'llamacpp';
+export type LocalRuntimeStatus = 'ready' | 'missing-model' | 'offline';
+
+export interface LocalRuntimeReadiness {
+  id: LocalRuntimeId;
+  name: string;
+  endpoint: string;
+  status: LocalRuntimeStatus;
+  commandHint: string;
+  recommendedModel: string | null;
+  recommendedModelInstalled: boolean;
+  detectedModelCount: number;
+}
+
+interface LocalRuntimeDefinition {
+  id: LocalRuntimeId;
+  name: string;
+  endpoint: string;
+  modelsUrl: string;
+  commandHint: string;
+  recommendedModel: string | null;
+  parseModelIds: (payload: unknown) => string[];
+}
+
+const LOCAL_RUNTIME_DEFINITIONS: LocalRuntimeDefinition[] = [
+  {
+    id: 'ollama',
+    name: 'Ollama',
+    endpoint: 'http://127.0.0.1:11434',
+    modelsUrl: 'http://127.0.0.1:11434/api/tags',
+    commandHint: 'ollama serve',
+    recommendedModel: 'gemma4:e4b',
+    parseModelIds: (payload) => {
+      const models = (payload as { models?: Array<{ name?: string }> })?.models;
+      return Array.isArray(models)
+        ? models
+            .map((model) => (typeof model?.name === 'string' ? model.name.trim() : ''))
+            .filter(Boolean)
+        : [];
+    },
+  },
+  {
+    id: 'llamacpp',
+    name: 'llama.cpp',
+    endpoint: 'http://127.0.0.1:8080',
+    modelsUrl: 'http://127.0.0.1:8080/v1/models',
+    commandHint: './server --host 0.0.0.0 --port 8080 -m model.gguf',
+    recommendedModel: null,
+    parseModelIds: (payload) => {
+      const models = (payload as { data?: Array<{ id?: string }> })?.data;
+      return Array.isArray(models)
+        ? models
+            .map((model) => (typeof model?.id === 'string' ? model.id.trim() : ''))
+            .filter(Boolean)
+        : [];
+    },
+  },
+];
+
+async function fetchModelIds(definition: LocalRuntimeDefinition): Promise<string[]> {
+  const response = await fetch(definition.modelsUrl, {
+    signal: AbortSignal.timeout(1500),
+  });
+
+  if (!response.ok) {
+    throw new Error(`${definition.id} readiness probe failed (${response.status})`);
+  }
+
+  return definition.parseModelIds(await response.json());
+}
+
+function toReadiness(
+  definition: LocalRuntimeDefinition,
+  modelIds: string[]
+): LocalRuntimeReadiness {
+  const recommendedModelInstalled = definition.recommendedModel
+    ? modelIds.some((modelId) => modelId === definition.recommendedModel)
+    : modelIds.length > 0;
+
+  const status: LocalRuntimeStatus =
+    modelIds.length === 0 || !recommendedModelInstalled ? 'missing-model' : 'ready';
+
+  return {
+    id: definition.id,
+    name: definition.name,
+    endpoint: definition.endpoint,
+    status,
+    commandHint:
+      status === 'missing-model' && definition.recommendedModel
+        ? `ollama pull ${definition.recommendedModel}`
+        : definition.commandHint,
+    recommendedModel: definition.recommendedModel,
+    recommendedModelInstalled,
+    detectedModelCount: modelIds.length,
+  };
+}
+
+function toOffline(definition: LocalRuntimeDefinition): LocalRuntimeReadiness {
+  return {
+    id: definition.id,
+    name: definition.name,
+    endpoint: definition.endpoint,
+    status: 'offline',
+    commandHint: definition.commandHint,
+    recommendedModel: definition.recommendedModel,
+    recommendedModelInstalled: false,
+    detectedModelCount: 0,
+  };
+}
+
+export async function getLocalRuntimeReadiness(): Promise<LocalRuntimeReadiness[]> {
+  return Promise.all(
+    LOCAL_RUNTIME_DEFINITIONS.map(async (definition) => {
+      try {
+        return toReadiness(definition, await fetchModelIds(definition));
+      } catch {
+        return toOffline(definition);
+      }
+    })
+  );
+}
diff --git a/src/web-server/routes/profile-routes.ts b/src/web-server/routes/profile-routes.ts
index d640d04e..84752b1d 100644
--- a/src/web-server/routes/profile-routes.ts
+++ b/src/web-server/routes/profile-routes.ts
@@ -10,6 +10,7 @@ import { isReservedName, RESERVED_PROFILE_NAMES } from '../../config/reserved-na
 import {
   createApiProfile,
   createCliproxyBridgeProfile,
+  getLocalRuntimeReadiness,
   removeApiProfile,
   updateApiProfileTarget,
   discoverApiProfileOrphans,
@@ -86,6 +87,14 @@ router.get('/cliproxy-bridge/providers', (_req: Request, res: Response): void =>
   }
 });
 
+router.get('/local-runtime-readiness', async (_req: Request, res: Response): Promise<void> => {
+  try {
+    res.json({ runtimes: await getLocalRuntimeReadiness() });
+  } catch (error) {
+    res.status(500).json({ error: (error as Error).message });
+  }
+});
+
 router.post('/cliproxy-bridge', (req: Request, res: Response): void => {
   const shape = validatePayloadShape(req.body, ['provider', 'name', 'target']);
   if (!shape.ok) {
diff --git a/tests/unit/web-server/profile-routes-local-runtime-readiness.test.ts b/tests/unit/web-server/profile-routes-local-runtime-readiness.test.ts
new file mode 100644
index 00000000..46c89b25
--- /dev/null
+++ b/tests/unit/web-server/profile-routes-local-runtime-readiness.test.ts
@@ -0,0 +1,141 @@
+import { afterAll, afterEach, beforeAll, beforeEach, describe, expect, it } from 'bun:test';
+import express from 'express';
+import type { Server } from 'http';
+import profileRoutes from '../../../src/web-server/routes/profile-routes';
+
+describe('profile-routes local runtime readiness', () => {
+  let server: Server;
+  let baseUrl = '';
+  const originalFetch = globalThis.fetch;
+
+  beforeAll(async () => {
+    const app = express();
+    app.use(express.json());
+    app.use('/api/profiles', profileRoutes);
+
+    await new Promise<void>((resolve, reject) => {
+      server = app.listen(0, '127.0.0.1');
+      const onError = (error: Error) => reject(error);
+      server.once('error', onError);
+      server.once('listening', () => {
+        server.off('error', onError);
+        resolve();
+      });
+    });
+
+    const address = server.address();
+    if (!address || typeof address === 'string') {
+      throw new Error('Unable to resolve test server port');
+    }
+
+    baseUrl = `http://127.0.0.1:${address.port}`;
+  });
+
+  afterAll(async () => {
+    await new Promise<void>((resolve) => server.close(() => resolve()));
+  });
+
+  beforeEach(() => {
+    globalThis.fetch = (async (input: RequestInfo | URL) => {
+      const url = String(input);
+
+      if (url.startsWith(baseUrl)) {
+        return originalFetch(input);
+      }
+
+      if (url.includes('11434/api/tags')) {
+        return new Response(
+          JSON.stringify({
+            models: [{ name: 'gemma4:e4b' }, { name: 'qwen3-coder:latest' }],
+          }),
+          {
+            status: 200,
+            headers: { 'Content-Type': 'application/json' },
+          }
+        );
+      }
+
+      if (url.includes('8080/v1/models')) {
+        return new Response(
+          JSON.stringify({
+            data: [{ id: 'Qwen3-Coder-30B-A3B-Instruct-Q4_K_M.gguf' }],
+          }),
+          {
+            status: 200,
+            headers: { 'Content-Type': 'application/json' },
+          }
+        );
+      }
+
+      throw new Error(`Unexpected URL: ${url}`);
+    }) as typeof fetch;
+  });
+
+  afterEach(() => {
+    globalThis.fetch = originalFetch;
+  });
+
+  it('reports local runtimes as ready when their endpoints respond with models', async () => {
+    const response = await fetch(`${baseUrl}/api/profiles/local-runtime-readiness`);
+
+    expect(response.status).toBe(200);
+    const body = (await response.json()) as {
+      runtimes: Array<{
+        id: string;
+        status: string;
+        recommendedModelInstalled: boolean;
+      }>;
+    };
+
+    expect(body.runtimes).toHaveLength(2);
+    expect(body.runtimes).toContainEqual(
+      expect.objectContaining({
+        id: 'ollama',
+        status: 'ready',
+        recommendedModelInstalled: true,
+      })
+    );
+    expect(body.runtimes).toContainEqual(
+      expect.objectContaining({
+        id: 'llamacpp',
+        status: 'ready',
+      })
+    );
+  });
+
+  it('reports setup guidance when local endpoints are unavailable', async () => {
+    globalThis.fetch = (async (input: RequestInfo | URL) => {
+      const url = String(input);
+      if (url.startsWith(baseUrl)) {
+        return originalFetch(input);
+      }
+      throw new Error('connect ECONNREFUSED');
+    }) as typeof fetch;
+
+    const response = await fetch(`${baseUrl}/api/profiles/local-runtime-readiness`);
+
+    expect(response.status).toBe(200);
+    const body = (await response.json()) as {
+      runtimes: Array<{
+        id: string;
+        status: string;
+        commandHint: string;
+      }>;
+    };
+
+    expect(body.runtimes).toContainEqual(
+      expect.objectContaining({
+        id: 'ollama',
+        status: 'offline',
+        commandHint: 'ollama serve',
+      })
+    );
+    expect(body.runtimes).toContainEqual(
+      expect.objectContaining({
+        id: 'llamacpp',
+        status: 'offline',
+        commandHint: './server --host 0.0.0.0 --port 8080 -m model.gguf',
+      })
+    );
+  });
+});
diff --git a/ui/src/hooks/use-local-runtime-readiness.ts b/ui/src/hooks/use-local-runtime-readiness.ts
new file mode 100644
index 00000000..c207ae58
--- /dev/null
+++ b/ui/src/hooks/use-local-runtime-readiness.ts
@@ -0,0 +1,11 @@
+import { useQuery } from '@tanstack/react-query';
+import { api } from '@/lib/api-client';
+
+export function useLocalRuntimeReadiness() {
+  return useQuery({
+    queryKey: ['profiles', 'local-runtime-readiness'],
+    queryFn: () => api.profiles.getLocalRuntimeReadiness(),
+    staleTime: 15_000,
+    refetchOnWindowFocus: false,
+  });
+}
diff --git a/ui/src/lib/api-client.ts b/ui/src/lib/api-client.ts
index d8493a99..74a848ee 100644
--- a/ui/src/lib/api-client.ts
+++ b/ui/src/lib/api-client.ts
@@ -239,6 +239,23 @@ export interface Profile {
   cliproxyBridge?: CliproxyBridgeMetadata | null;
 }
 
+export type LocalRuntimeStatus = 'ready' | 'missing-model' | 'offline';
+
+export interface LocalRuntimeReadiness {
+  id: 'ollama' | 'llamacpp';
+  name: string;
+  endpoint: string;
+  status: LocalRuntimeStatus;
+  commandHint: string;
+  recommendedModel: string | null;
+  recommendedModelInstalled: boolean;
+  detectedModelCount: number;
+}
+
+export interface LocalRuntimeReadinessResponse {
+  runtimes: LocalRuntimeReadiness[];
+}
+
 export interface CreateProfile {
   name: string;
   baseUrl: string;
@@ -1024,6 +1041,8 @@ export interface CliproxyRestartResult {
 export const api = {
   profiles: {
     list: () => request<{ profiles: Profile[] }>('/profiles'),
+    getLocalRuntimeReadiness: () =>
+      request<LocalRuntimeReadinessResponse>('/profiles/local-runtime-readiness'),
     create: (data: CreateProfile) =>
       request('/profiles', {
         method: 'POST',