[GH-ISSUE #1123] [BUG] 输出中莫名其妙的重复 confusing duplication in outputs

bsakaguchi commented

2026-10-26 21:26:24 +00:00

Owner

Originally created by @YMnRb on GitHub (Feb 25, 2026).
Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-V3/issues/1123

Describe the Bug

As mentioned in the title, it occasionally happens that, meaningless repeated content appears at the end of the output. It's somewhat similar to #1113 and #1033, but it seems not to be limited to situations involving complex programming logic problems as described in #1033.

大致如题所述，使用期间偶尔会出现，在输出内容的末尾出现无意义重复内容。和 #1113 有点像？也有点像 #1033，但好像不只是在涉及到复杂的编程逻辑问题（引自 #1033 ）时才会发生。

To Reproduce

This is an intermittent issue. And no stable reproduction method has been discovered.

偶发性问题，未发现稳定复现方法。

Expected Behaviour

Obviously, no such repetition is expected.

显然期望没有这样的重复。

Screenshots Examples (Chinese)

Example 1

[DeepSeek V3.2 response, 2069 tokens, 2026-01-18 16:16:51]
(以下内容为本次输出的末尾部分，前文表现正常，已略去)

3.  ***社交媒体时代的新解读***:
    *   Facebook/微信好友列表里大部分是典型的“弱连接”。它们让维持大量弱连接的代价变得极低。
    * “点赞之交”是数字时代的典型超弱链接形式。它们虽然极浅薄，但仍可能成为未来有价值连接的起点。
4.  
4.  
4.  
4.  
4.  

5.  
5.  

6.  

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.***
38.***
39.***
40.***
41.***
42.***
43.***
44.***
45.***
46.***

47.*
48.*
49.*
50.*
51.*
52.*
53.*
54.*
55.*

56.*

57.*

58.*

59.*

60.*

61.*

62.*

63.*

64.*

65.**

66.**

67.**

68.**

69.**

70.**

71.**

72.**

73.**

74.**

75.**

76.**

77.**

78.**

79.**

80.**

81**

82**

83**

84**

85**

86**

87**

88**

89**

90***

91***

92***

93***

94***

95***

96***

97***

98***

99***

[prompt, 2026-01-18 16:20:59]
第四条到第一百条是什么东西

[DeepSeek V3.2 response, 2026-01-18 16:20:59]

您观察得非常仔细！这确实是一个**非常奇怪且明显的错误**。

在之前的回答中，从“四、现实应用与启示”的第4点开始，后面出现了大量无意义的数字编号和星号，这**绝对不是**强连接/弱连接理论的内容。

这很可能是由于我在生成回答时出现了技术故障（例如，模型在输出过程中发生了“退化”或“循环”），或者我的内部处理模板出现了错误拼接。

(后略)

Example 2

Translated into English with the assistance from DeepL

（说句题外话，这些 issues 还有人看吗？）

Originally created by @YMnRb on GitHub (Feb 25, 2026). Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-V3/issues/1123 ## **Describe the Bug** As mentioned in the title, it occasionally happens that, meaningless repeated content appears at the end of the output. It's somewhat similar to #1113 and #1033, but it seems not to be limited to situations involving *complex programming logic problems* as described in #1033. 大致如题所述，使用期间偶尔会出现，在输出内容的末尾出现无意义重复内容。和 #1113 有点像？也有点像 #1033，但好像不只是在涉及到*复杂的编程逻辑问题*（引自 #1033 ）时才会发生。 ## **To Reproduce** This is an intermittent issue. And no stable reproduction method has been discovered. 偶发性问题，未发现稳定复现方法。 ## **Expected Behaviour** Obviously, no such repetition is expected. 显然期望没有这样的重复。 ## **~~Screenshots~~ Examples (Chinese)** <details><summary>Example 1</summary> > > `[DeepSeek V3.2 response, 2069 tokens, 2026-01-18 16:16:51]` > (以下内容为本次输出的末尾部分，前文表现正常，已略去) > > ``` > 3. ***社交媒体时代的新解读***: > * Facebook/微信好友列表里大部分是典型的“弱连接”。它们让维持大量弱连接的代价变得极低。 > * “点赞之交”是数字时代的典型超弱链接形式。它们虽然极浅薄，但仍可能成为未来有价值连接的起点。 > 4. > 4. > 4. > 4. > 4. > > 5. > 5. > > 6. > > 7. > > 8. > > 9. > > 10. > > 11. > > 12. > > 13. > > 14. > > 15. > > 16. > > 17. > > 18. > > 19. > > 20. > 21. > 22. > 23. > 24. > 25. > 26. > 27. > 28. > 29. > 30. > 31. > 32. > 33. > 34. > 35. > 36. > 37.*** > 38.*** > 39.*** > 40.*** > 41.*** > 42.*** > 43.*** > 44.*** > 45.*** > 46.*** > > 47.* > 48.* > 49.* > 50.* > 51.* > 52.* > 53.* > 54.* > 55.* > > 56.* > > 57.* > > 58.* > > 59.* > > 60.* > > 61.* > > 62.* > > 63.* > > 64.* > > 65.** > > 66.** > > 67.** > > 68.** > > 69.** > > 70.** > > 71.** > > 72.** > > 73.** > > 74.** > > 75.** > > 76.** > > 77.** > > 78.** > > 79.** > > 80.** > > 81** > > 82** > > 83** > > 84** > > 85** > > 86** > > 87** > > 88** > > 89** > > 90*** > > 91*** > > 92*** > > 93*** > > 94*** > > 95*** > > 96*** > > 97*** > > 98*** > > 99*** > > `[prompt, 2026-01-18 16:20:59]` > 第四条到第一百条是什么东西 > `[DeepSeek V3.2 response, 2026-01-18 16:20:59]` > ``` > 您观察得非常仔细！这确实是一个**非常奇怪且明显的错误**。 > > 在之前的回答中，从“四、现实应用与启示”的第4点开始，后面出现了大量无意义的数字编号和星号，这**绝对不是**强连接/弱连接理论的内容。 > > 这很可能是由于我在生成回答时出现了技术故障（例如，模型在输出过程中发生了“退化”或“循环”），或者我的内部处理模板出现了错误拼接。 > ``` > (后略) </details> <details> <summary>Example 2</summary> <img width="818" height="222" alt="Image: Example 2" src="https://github.com/user-attachments/assets/da47b390-b7f4-40be-92c6-1aeb84a867a9" /> </details> --- *Translated into English with the assistance from DeepL* （说句题外话，这些 issues 还有人看吗？）

bsakaguchi

2026-10-26 21:26:24 +00:00

closed this issue
added the
stale

closed-as-stale
labels

bsakaguchi commented

2026-10-26 21:26:26 +00:00

Author

Owner

@YMnRb commented on GitHub (Feb 25, 2026):

（题外话 2 ）另外感觉真该给这个 repo 开下 Discussion 板块的。莫名其妙的神人 issue 太多了。不如都赶到 Discussion 去。

@YMnRb commented on GitHub (Feb 25, 2026): （题外话 2 ）另外感觉真该给这个 repo 开下 Discussion 板块的。莫名其妙的神人 issue 太多了。不如都赶到 Discussion 去。

bsakaguchi commented

2026-10-26 21:26:26 +00:00

Author

Owner

@theBig-brother commented on GitHub (Mar 9, 2026):

不止你，我也有

@theBig-brother commented on GitHub (Mar 9, 2026): 不止你，我也有 <img width="1000" height="543" alt="Image" src="https://github.com/user-attachments/assets/d4e38494-24f4-4377-9789-6db7ae8c8d4c" /> <img width="683" height="616" alt="Image" src="https://github.com/user-attachments/assets/0c80c45b-c624-4c34-9bfe-0d761ad10d82" /> <img width="976" height="562" alt="Image" src="https://github.com/user-attachments/assets/787d0d87-5365-4649-a27e-1cc4c0dc0025" /> <img width="970" height="646" alt="Image" src="https://github.com/user-attachments/assets/12998732-28f3-4644-8d4f-7de4ff4b6771" />

bsakaguchi commented

2026-10-26 21:26:26 +00:00

Author

Owner

@github-actions[bot] commented on GitHub (Apr 9, 2026):

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!

@github-actions[bot] commented on GitHub (Apr 9, 2026): This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!

bsakaguchi commented

2026-10-26 21:26:26 +00:00

Author

Owner

@YMnRb commented on GitHub (Apr 12, 2026):

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!

@YMnRb commented on GitHub (Apr 12, 2026): > This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!

bsakaguchi commented

2026-10-26 21:26:26 +00:00

Author

Owner

@github-actions[bot] commented on GitHub (May 15, 2026):

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!

@github-actions[bot] commented on GitHub (May 15, 2026): This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!

bsakaguchi commented

2026-10-26 21:26:27 +00:00

Author

Owner

@github-actions[bot] commented on GitHub (May 30, 2026):

false

@github-actions[bot] commented on GitHub (May 30, 2026): false

Rows
Columns

[GH-ISSUE #1123] [BUG] 输出中莫名其妙的重复 confusing duplication in outputs #704

Describe the Bug

To Reproduce

Expected Behaviour

Screenshots Examples (Chinese)