diff --git a/packages/cli/src/ui/hooks/useVisionAutoSwitch.test.ts b/packages/cli/src/ui/hooks/useVisionAutoSwitch.test.ts
index 782986ce9..168f3a585 100644
--- a/packages/cli/src/ui/hooks/useVisionAutoSwitch.test.ts
+++ b/packages/cli/src/ui/hooks/useVisionAutoSwitch.test.ts
@@ -62,7 +62,7 @@ describe('useVisionAutoSwitch helpers', () => {
       const result = shouldOfferVisionSwitch(
         parts,
         AuthType.QWEN_OAUTH,
-        'vision-model',
+        'qwen3.5-plus',
         true,
       );
       expect(result).toBe(false);
@@ -161,7 +161,7 @@ describe('useVisionAutoSwitch helpers', () => {
       const result = shouldOfferVisionSwitch(
         parts,
         AuthType.QWEN_OAUTH,
-        'vision-model',
+        'qwen3.5-plus',
         true,
       );
       expect(result).toBe(false);
@@ -526,7 +526,7 @@ describe('useVisionAutoSwitch hook', () => {
     it('does not switch in YOLO mode when already using vision model', async () => {
       const config = createMockConfig(
         AuthType.QWEN_OAUTH,
-        'vision-model',
+        'qwen3.5-plus',
         ApprovalMode.YOLO,
       );
       const onVisionSwitchRequired = vi.fn();
@@ -728,7 +728,7 @@ describe('useVisionAutoSwitch hook', () => {
 
       expect(switchResult.shouldProceed).toBe(true);
       expect(switchResult.originalModel).toBe('qwen3-coder-plus');
-      expect(config.setModel).toHaveBeenCalledWith('vision-model', {
+      expect(config.setModel).toHaveBeenCalledWith('qwen3.5-plus', {
         reason: 'vision_auto_switch',
         context: 'Default VLM switch mode: once (one-time override)',
       });
@@ -764,7 +764,7 @@ describe('useVisionAutoSwitch hook', () => {
 
       expect(switchResult.shouldProceed).toBe(true);
       expect(switchResult.originalModel).toBeUndefined(); // No original model for session switch
-      expect(config.setModel).toHaveBeenCalledWith('vision-model', {
+      expect(config.setModel).toHaveBeenCalledWith('qwen3.5-plus', {
         reason: 'vision_auto_switch',
         context: 'Default VLM switch mode: session (session persistent)',
       });
@@ -813,7 +813,7 @@ describe('useVisionAutoSwitch hook', () => {
       );
       const onVisionSwitchRequired = vi
         .fn()
-        .mockResolvedValue({ modelOverride: 'vision-model' });
+        .mockResolvedValue({ modelOverride: 'qwen3.5-plus' });
       const { result } = renderHook(() =>
         useVisionAutoSwitch(
           config,
diff --git a/packages/cli/src/ui/models/availableModels.test.ts b/packages/cli/src/ui/models/availableModels.test.ts
index a17405d38..619ddbbdd 100644
--- a/packages/cli/src/ui/models/availableModels.test.ts
+++ b/packages/cli/src/ui/models/availableModels.test.ts
@@ -20,16 +20,23 @@ describe('availableModels', () => {
   describe('Qwen models', () => {
     const qwenModels = getFilteredQwenModels(true);
 
-    it('should include coder model', () => {
-      const coderModel = qwenModels.find((m) => m.id === MAINLINE_CODER);
+    it('should include default model (qwen3.5-plus)', () => {
+      const defaultModel = qwenModels.find((m) => m.id === MAINLINE_CODER);
+      expect(defaultModel).toBeDefined();
+      expect(defaultModel?.id).toBe('qwen3.5-plus');
+    });
+
+    it('should include coder-model as non-vision alternative', () => {
+      const coderModel = qwenModels.find((m) => m.id === 'coder-model');
       expect(coderModel).toBeDefined();
       expect(coderModel?.isVision).toBeFalsy();
     });
 
-    it('should include vision model', () => {
+    it('should include vision model (qwen3.5-plus with vision capability)', () => {
       const visionModel = qwenModels.find((m) => m.id === MAINLINE_VLM);
       expect(visionModel).toBeDefined();
       expect(visionModel?.isVision).toBe(true);
+      expect(visionModel?.id).toBe('qwen3.5-plus');
     });
   });
 
@@ -197,8 +204,8 @@ describe('availableModels', () => {
       expect(isVisionModel(MAINLINE_VLM)).toBe(true);
     });
 
-    it('should return false for non-vision model', () => {
-      expect(isVisionModel(MAINLINE_CODER)).toBe(false);
+    it('should return false for coder-model (non-vision)', () => {
+      expect(isVisionModel('coder-model')).toBe(false);
     });
 
     it('should return false for unknown model', () => {
diff --git a/packages/cli/src/ui/models/availableModels.ts b/packages/cli/src/ui/models/availableModels.ts
index a68cc69cc..54f36cae1 100644
--- a/packages/cli/src/ui/models/availableModels.ts
+++ b/packages/cli/src/ui/models/availableModels.ts
@@ -20,7 +20,7 @@ export type AvailableModel = {
   isVision?: boolean;
 };
 
-export const MAINLINE_VLM = 'vision-model';
+export const MAINLINE_VLM = 'qwen3.5-plus';
 export const MAINLINE_CODER = DEFAULT_QWEN_MODEL;
 
 const CACHED_QWEN_OAUTH_MODELS: AvailableModel[] = QWEN_OAUTH_MODELS.map(
diff --git a/packages/core/src/config/config.test.ts b/packages/core/src/config/config.test.ts
index 4d2af04b3..9126c0983 100644
--- a/packages/core/src/config/config.test.ts
+++ b/packages/core/src/config/config.test.ts
@@ -370,9 +370,9 @@ describe('Server Config (config.ts)', () => {
       // Spy after initial refresh to ensure model switch does not re-trigger refreshAuth.
       const refreshSpy = vi.spyOn(config, 'refreshAuth');
 
-      await config.switchModel(AuthType.QWEN_OAUTH, 'vision-model');
+      await config.switchModel(AuthType.QWEN_OAUTH, 'qwen3.5-plus');
 
-      expect(config.getModel()).toBe('vision-model');
+      expect(config.getModel()).toBe('qwen3.5-plus');
       expect(refreshSpy).not.toHaveBeenCalled();
       // Called once during initial refreshAuth + once during handleModelChange diffing.
       expect(
diff --git a/packages/core/src/core/contentGenerator.test.ts b/packages/core/src/core/contentGenerator.test.ts
index eef7f5ac8..312af8069 100644
--- a/packages/core/src/core/contentGenerator.test.ts
+++ b/packages/core/src/core/contentGenerator.test.ts
@@ -90,11 +90,11 @@ describe('createContentGeneratorConfig', () => {
 
   it('should preserve provided fields and set authType for QWEN_OAUTH', () => {
     const cfg = createContentGeneratorConfig(mockConfig, AuthType.QWEN_OAUTH, {
-      model: 'vision-model',
+      model: 'qwen3.5-plus',
       apiKey: 'QWEN_OAUTH_DYNAMIC_TOKEN',
     });
     expect(cfg.authType).toBe(AuthType.QWEN_OAUTH);
-    expect(cfg.model).toBe('vision-model');
+    expect(cfg.model).toBe('qwen3.5-plus');
     expect(cfg.apiKey).toBe('QWEN_OAUTH_DYNAMIC_TOKEN');
   });
 
diff --git a/packages/core/src/core/openaiContentGenerator/provider/dashscope.test.ts b/packages/core/src/core/openaiContentGenerator/provider/dashscope.test.ts
index a57bbacb7..f73e2c54b 100644
--- a/packages/core/src/core/openaiContentGenerator/provider/dashscope.test.ts
+++ b/packages/core/src/core/openaiContentGenerator/provider/dashscope.test.ts
@@ -884,9 +884,9 @@ describe('DashScopeOpenAICompatibleProvider', () => {
       ).toBe(true);
     });
 
-    it('should set high resolution flag for the vision-model alias', () => {
+    it('should set high resolution flag for the qwen3.5-plus model', () => {
       const request: OpenAI.Chat.ChatCompletionCreateParams = {
-        model: 'vision-model',
+        model: 'qwen3.5-plus',
         messages: [
           {
             role: 'user',
diff --git a/packages/core/src/core/openaiContentGenerator/provider/dashscope.ts b/packages/core/src/core/openaiContentGenerator/provider/dashscope.ts
index ccf201e24..c1c4d0536 100644
--- a/packages/core/src/core/openaiContentGenerator/provider/dashscope.ts
+++ b/packages/core/src/core/openaiContentGenerator/provider/dashscope.ts
@@ -286,7 +286,7 @@ export class DashScopeOpenAICompatibleProvider
 
     const normalized = model.toLowerCase();
 
-    if (normalized === 'vision-model') {
+    if (normalized === 'qwen3.5-plus') {
       return true;
     }
 
diff --git a/packages/core/src/core/prompts.ts b/packages/core/src/core/prompts.ts
index 60c2c56d9..fd321e7ff 100644
--- a/packages/core/src/core/prompts.ts
+++ b/packages/core/src/core/prompts.ts
@@ -801,9 +801,9 @@ function getToolCallExamples(model?: string): string {
     if (/coder-model/i.test(model)) {
       return qwenCoderToolCallExamples;
     }
-    // Match vision-model pattern (same as qwen3-vl)
-    if (/vision-model/i.test(model)) {
-      return qwenVlToolCallExamples;
+    // Match qwen3.5-plus pattern (uses same tool call examples as coder)
+    if (/qwen3\.5-plus/i.test(model)) {
+      return qwenCoderToolCallExamples;
     }
   }
 
diff --git a/packages/core/src/core/tokenLimits.ts b/packages/core/src/core/tokenLimits.ts
index e65734362..d88dd28f7 100644
--- a/packages/core/src/core/tokenLimits.ts
+++ b/packages/core/src/core/tokenLimits.ts
@@ -150,9 +150,6 @@ const PATTERNS: Array<[RegExp, TokenCount]> = [
   [/^qwen3-vl-plus$/, LIMITS['256k']], // Qwen3-VL-Plus: 256K input
   [/^qwen-vl-max.*$/, LIMITS['128k']],
 
-  // Generic vision-model: same as qwen-vl-max (128K token context)
-  [/^vision-model$/, LIMITS['128k']],
-
   // -------------------
   // ByteDance Seed-OSS (512K)
   // -------------------
@@ -216,9 +213,6 @@ const OUTPUT_PATTERNS: Array<[RegExp, TokenCount]> = [
   // Qwen-VL-Max-Latest: 8,192 max output tokens
   [/^qwen-vl-max-latest$/, LIMITS['8k']],
 
-  // Generic vision-model: same as qwen-vl-max-latest (8K max output tokens)
-  [/^vision-model$/, LIMITS['8k']],
-
   // Qwen3-VL-Plus: 32K max output tokens
   [/^qwen3-vl-plus$/, LIMITS['32k']],
 
diff --git a/packages/core/src/models/constants.ts b/packages/core/src/models/constants.ts
index c4c72d9e5..f918e3691 100644
--- a/packages/core/src/models/constants.ts
+++ b/packages/core/src/models/constants.ts
@@ -102,7 +102,7 @@ export const QWEN_OAUTH_MODELS: ModelConfig[] = [
     name: 'qwen3.5-plus',
     description:
       'The Qwen3.5 native vision-language series Plus models deliver outstanding performance comparable to the very latest state-of-the-art models, with significant leaps in both pure-text and multimodal capabilities compared to the 3 series.',
-    capabilities: { vision: false },
+    capabilities: { vision: true },
   },
   {
     id: 'coder-model',
@@ -111,12 +111,6 @@ export const QWEN_OAUTH_MODELS: ModelConfig[] = [
       'Qwen 3.5 Plus — efficient hybrid model with leading coding performance',
     capabilities: { vision: false },
   },
-  {
-    id: 'vision-model',
-    name: 'vision-model',
-    description: 'The latest Qwen Vision model from Alibaba Cloud ModelStudio',
-    capabilities: { vision: true },
-  },
 ];
 
 /**
diff --git a/packages/core/src/models/modelConfigResolver.test.ts b/packages/core/src/models/modelConfigResolver.test.ts
index b045bcbdd..bc31e81e5 100644
--- a/packages/core/src/models/modelConfigResolver.test.ts
+++ b/packages/core/src/models/modelConfigResolver.test.ts
@@ -157,17 +157,17 @@ describe('modelConfigResolver', () => {
         expect(result.sources['apiKey'].kind).toBe('computed');
       });
 
-      it('allows vision-model for Qwen OAuth', () => {
+      it('allows qwen3.5-plus for Qwen OAuth', () => {
         const result = resolveModelConfig({
           authType: AuthType.QWEN_OAUTH,
           cli: {
-            model: 'vision-model',
+            model: 'qwen3.5-plus',
           },
           settings: {},
           env: {},
         });
 
-        expect(result.config.model).toBe('vision-model');
+        expect(result.config.model).toBe('qwen3.5-plus');
         expect(result.sources['model'].kind).toBe('cli');
       });
 
diff --git a/packages/core/src/models/modelRegistry.test.ts b/packages/core/src/models/modelRegistry.test.ts
index 67510d5ec..f3c08ec44 100644
--- a/packages/core/src/models/modelRegistry.test.ts
+++ b/packages/core/src/models/modelRegistry.test.ts
@@ -18,7 +18,6 @@ describe('ModelRegistry', () => {
       expect(qwenModels.length).toBe(QWEN_OAUTH_MODELS.length);
       expect(qwenModels[0].id).toBe('qwen3.5-plus');
       expect(qwenModels[1].id).toBe('coder-model');
-      expect(qwenModels[2].id).toBe('vision-model');
     });
 
     it('should initialize with empty config', () => {
diff --git a/packages/core/src/models/modelsConfig.test.ts b/packages/core/src/models/modelsConfig.test.ts
index 7215d3905..64c54b0ab 100644
--- a/packages/core/src/models/modelsConfig.test.ts
+++ b/packages/core/src/models/modelsConfig.test.ts
@@ -454,7 +454,7 @@ describe('ModelsConfig', () => {
     });
 
     // Switching within qwen-oauth triggers applyResolvedModelDefaults().
-    await modelsConfig.switchModel(AuthType.QWEN_OAUTH, 'vision-model');
+    await modelsConfig.switchModel(AuthType.QWEN_OAUTH, 'qwen3.5-plus');
 
     const gc = currentGenerationConfig(modelsConfig);
     expect(gc.apiKey).toBe('QWEN_OAUTH_DYNAMIC_TOKEN');
diff --git a/packages/core/src/models/modelsConfig.ts b/packages/core/src/models/modelsConfig.ts
index 9311c9279..ff4543746 100644
--- a/packages/core/src/models/modelsConfig.ts
+++ b/packages/core/src/models/modelsConfig.ts
@@ -309,9 +309,10 @@ export class ModelsConfig {
     metadata?: ModelSwitchMetadata,
   ): Promise<void> {
     // Special case: qwen-oauth VLM auto-switch - hot update in place
+    // Support hot-update for both qwen3.5-plus (vision-capable) and coder-model
     if (
       this.currentAuthType === AuthType.QWEN_OAUTH &&
-      (newModel === DEFAULT_QWEN_MODEL || newModel === 'vision-model')
+      (newModel === 'qwen3.5-plus' || newModel === 'coder-model')
     ) {
       this.strictModelProviderSelection = false;
       this._generationConfig.model = newModel;
@@ -782,7 +783,7 @@ export class ModelsConfig {
    * - We're checking if switching between two models within the SAME authType needs refresh
    *
    * Examples:
-   * - Qwen OAuth: coder-model -> vision-model (same authType, hot-update safe)
+   * - Qwen OAuth: coder-model -> qwen3.5-plus (same authType, hot-update safe)
    * - OpenAI: model-a -> model-b with same envKey (same authType, hot-update safe)
    * - OpenAI: gpt-4 -> deepseek-chat with different envKey (same authType, needs refresh)
    *
@@ -799,7 +800,7 @@ export class ModelsConfig {
     }
 
     // For Qwen OAuth, model switches within the same authType can always be hot-updated
-    // (coder-model <-> vision-model don't require ContentGenerator recreation)
+    // (coder-model <-> qwen3.5-plus don't require ContentGenerator recreation)
     if (authType === AuthType.QWEN_OAUTH) {
       return false;
     }