diff --git a/orchestrator/src/client/pages/SettingsPage.test.tsx b/orchestrator/src/client/pages/SettingsPage.test.tsx
index ba334f3..93bdd86 100644
--- a/orchestrator/src/client/pages/SettingsPage.test.tsx
+++ b/orchestrator/src/client/pages/SettingsPage.test.tsx
@@ -23,14 +23,14 @@ vi.mock("sonner", () => ({
 }))
 
 const baseSettings: AppSettings = {
-  model: "openai/gpt-4o-mini",
-  defaultModel: "openai/gpt-4o-mini",
+  model: "google/gemini-3-flash-preview",
+  defaultModel: "google/gemini-3-flash-preview",
   overrideModel: null,
-  modelScorer: "openai/gpt-4o-mini",
+  modelScorer: "google/gemini-3-flash-preview",
   overrideModelScorer: null,
-  modelTailoring: "openai/gpt-4o-mini",
+  modelTailoring: "google/gemini-3-flash-preview",
   overrideModelTailoring: null,
-  modelProjectSelection: "openai/gpt-4o-mini",
+  modelProjectSelection: "google/gemini-3-flash-preview",
   overrideModelProjectSelection: null,
   pipelineWebhookUrl: "",
   defaultPipelineWebhookUrl: "",
@@ -160,7 +160,7 @@ describe("SettingsPage", () => {
 
     const modelField = screen.getByText("Override model").parentElement ?? screen.getByRole("main")
     const modelInput = within(modelField).getByRole("textbox")
-    
+
     // Change to > 200 chars
     fireEvent.change(modelInput, { target: { value: "a".repeat(201) } })
 
diff --git a/orchestrator/src/client/pages/settings/components/ModelSettingsSection.tsx b/orchestrator/src/client/pages/settings/components/ModelSettingsSection.tsx
index 7632ee2..29a9297 100644
--- a/orchestrator/src/client/pages/settings/components/ModelSettingsSection.tsx
+++ b/orchestrator/src/client/pages/settings/components/ModelSettingsSection.tsx
@@ -31,7 +31,7 @@ export const ModelSettingsSection: React.FC<ModelSettingsSectionProps> = ({
           <SettingsInput
             label="Override model"
             inputProps={register("model")}
-            placeholder={defaultModel || "openai/gpt-4o-mini"}
+            placeholder={defaultModel || "google/gemini-3-flash-preview"}
             disabled={isLoading || isSaving}
             error={errors.model?.message as string | undefined}
             helper="Leave blank to use the default from server env (`MODEL`)."
diff --git a/orchestrator/src/server/services/manualJob.ts b/orchestrator/src/server/services/manualJob.ts
index 7ceb079..a4dc149 100644
--- a/orchestrator/src/server/services/manualJob.ts
+++ b/orchestrator/src/server/services/manualJob.ts
@@ -62,7 +62,7 @@ export async function inferManualJobDetails(jobDescription: string): Promise<Man
   }
 
   const overrideModel = await getSetting('model');
-  const model = overrideModel || process.env.MODEL || 'openai/gpt-4o-mini';
+  const model = overrideModel || process.env.MODEL || 'google/gemini-3-flash-preview';
   const prompt = buildInferencePrompt(jobDescription);
 
   const result = await callOpenRouter<ManualJobApiResponse>({
diff --git a/orchestrator/src/server/services/projectSelection.ts b/orchestrator/src/server/services/projectSelection.ts
index d78d56b..36c63fc 100644
--- a/orchestrator/src/server/services/projectSelection.ts
+++ b/orchestrator/src/server/services/projectSelection.ts
@@ -43,7 +43,7 @@ export async function pickProjectIdsForJob(args: {
     getSetting('modelProjectSelection'),
   ]);
   // Precedence: Project-specific override > Global override > Env var > Default
-  const model = overrideModelProjectSelection || overrideModel || process.env.MODEL || 'openai/gpt-4o-mini';
+  const model = overrideModelProjectSelection || overrideModel || process.env.MODEL || 'google/gemini-3-flash-preview';
 
   const prompt = buildProjectSelectionPrompt({
     jobDescription: args.jobDescription,
diff --git a/orchestrator/src/server/services/scorer.ts b/orchestrator/src/server/services/scorer.ts
index e6a09e8..7c64bdb 100644
--- a/orchestrator/src/server/services/scorer.ts
+++ b/orchestrator/src/server/services/scorer.ts
@@ -49,7 +49,7 @@ export async function scoreJobSuitability(
     getSetting('modelScorer'),
   ]);
   // Precedence: Scorer-specific override > Global override > Env var > Default
-  const model = overrideModelScorer || overrideModel || process.env.MODEL || 'openai/gpt-4o-mini';
+  const model = overrideModelScorer || overrideModel || process.env.MODEL || 'google/gemini-3-flash-preview';
 
   const prompt = buildScoringPrompt(job, profile);
 
diff --git a/orchestrator/src/server/services/settings.ts b/orchestrator/src/server/services/settings.ts
index 76e47b1..907f6ce 100644
--- a/orchestrator/src/server/services/settings.ts
+++ b/orchestrator/src/server/services/settings.ts
@@ -19,7 +19,7 @@ export async function getEffectiveSettings(): Promise<AppSettings> {
 
   const envSettings = await getEnvSettingsData(overrides);
 
-  const defaultModel = process.env.MODEL || 'openai/gpt-4o-mini';
+  const defaultModel = process.env.MODEL || 'google/gemini-3-flash-preview';
   const overrideModel = overrides.model ?? null;
   const model = overrideModel || defaultModel;
 
diff --git a/orchestrator/src/server/services/summary.ts b/orchestrator/src/server/services/summary.ts
index d339a78..9a9e27b 100644
--- a/orchestrator/src/server/services/summary.ts
+++ b/orchestrator/src/server/services/summary.ts
@@ -74,7 +74,7 @@ export async function generateTailoring(
     getSetting('modelTailoring'),
   ]);
   // Precedence: Tailoring-specific override > Global override > Env var > Default
-  const model = overrideModelTailoring || overrideModel || process.env.MODEL || 'openai/gpt-4o-mini';
+  const model = overrideModelTailoring || overrideModel || process.env.MODEL || 'google/gemini-3-flash-preview';
   const prompt = buildTailoringPrompt(profile, jobDescription);
 
   const result = await callOpenRouter<TailoredData>({
diff --git a/resume-generator/generate_summary.py b/resume-generator/generate_summary.py
index a6c7ea0..0c12065 100644
--- a/resume-generator/generate_summary.py
+++ b/resume-generator/generate_summary.py
@@ -110,7 +110,7 @@ def generate_resume_summary(
     load_dotenv()
 
     api_key = os.getenv("OPENROUTER_API_KEY")
-    model = os.getenv("MODEL", "openai/gpt-4o-mini")
+    model = os.getenv("MODEL", "google/gemini-3-flash-preview")
 
     if not api_key:
         raise RuntimeError("Missing OPENROUTER_API_KEY in .env")