diff --git a/packages/core/src/core/openaiContentGenerator/provider/dashscope.test.ts b/packages/core/src/core/openaiContentGenerator/provider/dashscope.test.ts
index e1ecb61b6..024e9a28c 100644
--- a/packages/core/src/core/openaiContentGenerator/provider/dashscope.test.ts
+++ b/packages/core/src/core/openaiContentGenerator/provider/dashscope.test.ts
@@ -817,12 +817,12 @@ describe('DashScopeOpenAICompatibleProvider', () => {
       const request: OpenAI.Chat.ChatCompletionCreateParams = {
         model: 'unknown-model',
         messages: [{ role: 'user', content: 'Hello' }],
-        max_tokens: 10000, // Exceeds the default limit
+        max_tokens: 20000, // Exceeds the default limit
       };
 
       const result = provider.buildRequest(request, 'test-prompt-id');
 
-      expect(result.max_tokens).toBe(8192); // Should be limited to default output limit (8K)
+      expect(result.max_tokens).toBe(16384); // Should be limited to default output limit (16K)
     });
 
     it('should preserve other request parameters when limiting max_tokens', () => {
diff --git a/packages/core/src/core/tokenLimits.ts b/packages/core/src/core/tokenLimits.ts
index 2807e56c1..b566a01dc 100644
--- a/packages/core/src/core/tokenLimits.ts
+++ b/packages/core/src/core/tokenLimits.ts
@@ -9,7 +9,7 @@ type TokenCount = number;
 export type TokenLimitType = 'input' | 'output';
 
 export const DEFAULT_TOKEN_LIMIT: TokenCount = 131_072; // 128K (power-of-two)
-export const DEFAULT_OUTPUT_TOKEN_LIMIT: TokenCount = 8_192; // 8K tokens
+export const DEFAULT_OUTPUT_TOKEN_LIMIT: TokenCount = 16_384; // 16K tokens
 
 /**
  * Accurate numeric limits:
@@ -166,6 +166,7 @@ const OUTPUT_PATTERNS: Array<[RegExp, TokenCount]> = [
   [/^qwen3\.5/, LIMITS['64k']],
   [/^coder-model$/, LIMITS['64k']],
   [/^qwen3-max/, LIMITS['64k']],
+  [/^qwen/, LIMITS['8k']], // Qwen fallback (VL, turbo, plus, etc.): 8K
 
   // DeepSeek
   [/^deepseek-reasoner/, LIMITS['64k']],