florianvoss commited on
Commit
0855a5a
·
verified ·
1 Parent(s): 1d1a8cf

Upload compiled artifacts from Qwen3-VL-4B-Instruct

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +176 -0
  2. devkit/Qwen3-VL-4B-Instruct_language_embeddings.npy +3 -0
  3. devkit/chat_template.json +4 -0
  4. devkit/config.json +63 -0
  5. devkit/generation_config.json +14 -0
  6. devkit/merges.txt +0 -0
  7. devkit/precision.json +882 -0
  8. devkit/preprocessor_config.json +21 -0
  9. devkit/tokenizer.json +0 -0
  10. devkit/tokenizer_config.json +239 -0
  11. devkit/video_preprocessor_config.json +21 -0
  12. devkit/vlm_config.json +159 -0
  13. devkit/vocab.json +0 -0
  14. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token0_stage1_mla.elf +3 -0
  15. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1024_stage1_mla.elf +3 -0
  16. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1152_stage1_mla.elf +3 -0
  17. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1280_stage1_mla.elf +3 -0
  18. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token128_stage1_mla.elf +3 -0
  19. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1408_stage1_mla.elf +3 -0
  20. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1536_stage1_mla.elf +3 -0
  21. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1664_stage1_mla.elf +3 -0
  22. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1792_stage1_mla.elf +3 -0
  23. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1920_stage1_mla.elf +3 -0
  24. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token256_stage1_mla.elf +3 -0
  25. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token384_stage1_mla.elf +3 -0
  26. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token512_stage1_mla.elf +3 -0
  27. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token640_stage1_mla.elf +3 -0
  28. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token768_stage1_mla.elf +3 -0
  29. elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token896_stage1_mla.elf +3 -0
  30. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer0_stage1_mla.elf +3 -0
  31. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer10_stage1_mla.elf +3 -0
  32. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer11_stage1_mla.elf +3 -0
  33. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer12_stage1_mla.elf +3 -0
  34. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer13_stage1_mla.elf +3 -0
  35. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer14_stage1_mla.elf +3 -0
  36. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer15_stage1_mla.elf +3 -0
  37. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer16_stage1_mla.elf +3 -0
  38. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer17_stage1_mla.elf +3 -0
  39. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer18_stage1_mla.elf +3 -0
  40. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer19_stage1_mla.elf +3 -0
  41. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer1_stage1_mla.elf +3 -0
  42. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer20_stage1_mla.elf +3 -0
  43. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer21_stage1_mla.elf +3 -0
  44. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer22_stage1_mla.elf +3 -0
  45. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer23_stage1_mla.elf +3 -0
  46. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer24_stage1_mla.elf +3 -0
  47. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer25_stage1_mla.elf +3 -0
  48. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer26_stage1_mla.elf +3 -0
  49. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer27_stage1_mla.elf +3 -0
  50. elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer28_stage1_mla.elf +3 -0
.gitattributes CHANGED
@@ -33,3 +33,179 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token0_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
37
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1024_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
38
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1152_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
39
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1280_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
40
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token128_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
41
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1408_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
42
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1536_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
43
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1664_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
44
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1792_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
45
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1920_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
46
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token256_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
47
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token384_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
48
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token512_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
49
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token640_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
50
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token768_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
51
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token896_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
52
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer0_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
53
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer10_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
54
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer11_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
55
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer12_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
56
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer13_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
57
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer14_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
58
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer15_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
59
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer16_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
60
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer17_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
61
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer18_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
62
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer19_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
63
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer1_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
64
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer20_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
65
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer21_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
66
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer22_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
67
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer23_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
68
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer24_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
69
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer25_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
70
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer26_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
71
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer27_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
72
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer28_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
73
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer29_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
74
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer2_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
75
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer30_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
76
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer31_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
77
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer32_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
78
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer33_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
79
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer34_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
80
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer3_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
81
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer4_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
82
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer5_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
83
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer6_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
84
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer7_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
85
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer8_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
86
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer9_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
87
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer0_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
88
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer10_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
89
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer11_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
90
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer12_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
91
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer13_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
92
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer14_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
93
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer15_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
94
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer16_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
95
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer17_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
96
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer18_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
97
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer19_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
98
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer1_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
99
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer20_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
100
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer21_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
101
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer22_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
102
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer23_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
103
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer24_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
104
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer25_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
105
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer26_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
106
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer27_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
107
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer28_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
108
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer29_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
109
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer2_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
110
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer30_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
111
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer31_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
112
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer32_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
113
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer33_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
114
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer34_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
115
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer35_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
116
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer3_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
117
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer4_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
118
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer5_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
119
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer6_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
120
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer7_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
121
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer8_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
122
+ elf_files/Qwen3-VL-4B-Instruct_language_n128_pre_layer9_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
123
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token1023_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
124
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token1151_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
125
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token1279_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
126
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token127_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
127
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token1407_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
128
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token1535_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
129
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token1663_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
130
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token1791_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
131
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token1919_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
132
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token2047_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
133
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token255_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
134
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token383_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
135
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token511_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
136
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token639_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
137
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token767_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
138
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_cache_token895_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
139
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer0_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
140
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer10_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
141
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer11_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
142
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer12_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
143
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer13_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
144
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer14_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
145
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer15_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
146
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer16_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
147
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer17_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
148
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer18_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
149
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer19_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
150
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer1_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
151
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer20_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
152
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer21_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
153
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer22_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
154
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer23_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
155
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer24_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
156
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer25_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
157
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer26_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
158
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer27_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
159
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer28_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
160
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer29_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
161
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer2_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
162
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer30_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
163
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer31_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
164
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer32_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
165
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer33_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
166
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer34_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
167
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer35_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
168
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer3_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
169
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer4_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
170
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer5_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
171
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer6_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
172
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer7_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
173
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer8_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
174
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_post_layer9_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
175
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer0_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
176
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer10_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
177
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer11_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
178
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer12_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
179
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer13_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
180
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer14_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
181
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer15_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
182
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer16_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
183
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer17_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
184
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer18_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
185
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer19_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
186
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer1_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
187
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer20_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
188
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer21_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
189
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer22_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
190
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer23_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
191
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer24_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
192
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer25_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
193
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer26_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
194
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer27_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
195
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer28_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
196
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer29_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
197
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer2_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
198
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer30_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
199
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer31_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
200
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer32_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
201
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer33_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
202
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer34_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
203
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer35_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
204
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer3_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
205
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer4_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
206
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer5_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
207
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer6_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
208
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer7_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
209
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer8_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
210
+ elf_files/Qwen3-VL-4B-Instruct_language_n1_pre_layer9_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
211
+ elf_files/Qwen3-VL-4B-Instruct_vision_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text
devkit/Qwen3-VL-4B-Instruct_language_embeddings.npy ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e35cb036ea8f2c2dd3333a69035171111ee8dece6bddf48a6b07d0df4046aea9
3
+ size 777912448
devkit/chat_template.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0].role == 'system' %}\n {%- if messages[0].content is string %}\n {{- messages[0].content }}\n {%- else %}\n {%- for content in messages[0].content %}\n {%- if 'text' in content %}\n {{- content.text }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {{- '\\n\\n' }}\n {%- endif %}\n {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0].content is string %}\n {{- messages[0].content }}\n {%- else %}\n {%- for content in messages[0].content %}\n {%- if 'text' in content %}\n {{- content.text }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- set image_count = namespace(value=0) %}\n{%- set video_count = namespace(value=0) %}\n{%- for message in messages %}\n {%- if message.role == \"user\" %}\n {{- '<|im_start|>' + message.role + '\\n' }}\n {%- if message.content is string %}\n {{- message.content }}\n {%- else %}\n {%- for content in message.content %}\n {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}\n {%- set image_count.value = image_count.value + 1 %}\n {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}\n <|vision_start|><|image_pad|><|vision_end|>\n {%- elif content.type == 'video' or 'video' in content %}\n {%- set video_count.value = video_count.value + 1 %}\n {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}\n <|vision_start|><|video_pad|><|vision_end|>\n {%- elif 'text' in content %}\n {{- content.text }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role + '\\n' }}\n {%- if message.content is string %}\n {{- message.content }}\n {%- else %}\n {%- for content_item in message.content %}\n {%- if 'text' in content_item %}\n {{- content_item.text }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {%- if message.tool_calls %}\n {%- for tool_call in message.tool_calls %}\n {%- if (loop.first and message.content) or (not loop.first) %}\n {{- '\\n' }}\n {%- endif %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {%- if tool_call.arguments is string %}\n {{- tool_call.arguments }}\n {%- else %}\n {{- tool_call.arguments | tojson }}\n {%- endif %}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {%- if message.content is string %}\n {{- message.content }}\n {%- else %}\n {%- for content in message.content %}\n {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}\n {%- set image_count.value = image_count.value + 1 %}\n {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}\n <|vision_start|><|image_pad|><|vision_end|>\n {%- elif content.type == 'video' or 'video' in content %}\n {%- set video_count.value = video_count.value + 1 %}\n {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}\n <|vision_start|><|video_pad|><|vision_end|>\n {%- elif 'text' in content %}\n {{- content.text }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n"
3
+ }
4
+
devkit/config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3VLForConditionalGeneration"
4
+ ],
5
+ "image_token_id": 151655,
6
+ "model_type": "qwen3_vl",
7
+ "text_config": {
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "bos_token_id": 151643,
11
+ "dtype": "bfloat16",
12
+ "eos_token_id": 151645,
13
+ "head_dim": 128,
14
+ "hidden_act": "silu",
15
+ "hidden_size": 2560,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 9728,
18
+ "max_position_embeddings": 262144,
19
+ "model_type": "qwen3_vl_text",
20
+ "num_attention_heads": 32,
21
+ "num_hidden_layers": 36,
22
+ "num_key_value_heads": 8,
23
+ "rms_norm_eps": 1e-06,
24
+ "rope_scaling": {
25
+ "mrope_interleaved": true,
26
+ "mrope_section": [
27
+ 24,
28
+ 20,
29
+ 20
30
+ ],
31
+ "rope_type": "default"
32
+ },
33
+ "rope_theta": 5000000,
34
+ "tie_word_embeddings": true,
35
+ "use_cache": true,
36
+ "vocab_size": 151936
37
+ },
38
+ "tie_word_embeddings": true,
39
+ "transformers_version": "4.57.0.dev0",
40
+ "video_token_id": 151656,
41
+ "vision_config": {
42
+ "deepstack_visual_indexes": [
43
+ 5,
44
+ 11,
45
+ 17
46
+ ],
47
+ "depth": 24,
48
+ "hidden_act": "gelu_pytorch_tanh",
49
+ "hidden_size": 1024,
50
+ "in_channels": 3,
51
+ "initializer_range": 0.02,
52
+ "intermediate_size": 4096,
53
+ "model_type": "qwen3_vl",
54
+ "num_heads": 16,
55
+ "num_position_embeddings": 2304,
56
+ "out_hidden_size": 2560,
57
+ "patch_size": 16,
58
+ "spatial_merge_size": 2,
59
+ "temporal_patch_size": 2
60
+ },
61
+ "vision_end_token_id": 151653,
62
+ "vision_start_token_id": 151652
63
+ }
devkit/generation_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "pad_token_id": 151643,
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 151645,
7
+ 151643
8
+ ],
9
+ "top_k": 20,
10
+ "top_p": 0.8,
11
+ "repetition_penalty": 1.0,
12
+ "temperature": 0.7,
13
+ "transformers_version": "4.56.0"
14
+ }
devkit/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
devkit/precision.json ADDED
@@ -0,0 +1,882 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "part": "group_pre",
4
+ "idx": 0,
5
+ "precision": "A_BF16_W_INT8"
6
+ },
7
+ {
8
+ "part": "group_post",
9
+ "idx": 0,
10
+ "precision": "A_BF16_W_INT8"
11
+ },
12
+ {
13
+ "part": "single_pre",
14
+ "idx": 0,
15
+ "precision": "A_BF16_W_INT4"
16
+ },
17
+ {
18
+ "part": "single_post",
19
+ "idx": 0,
20
+ "precision": "A_BF16_W_INT4"
21
+ },
22
+ {
23
+ "part": "group_pre",
24
+ "idx": 1,
25
+ "precision": "A_BF16_W_INT8"
26
+ },
27
+ {
28
+ "part": "group_post",
29
+ "idx": 1,
30
+ "precision": "A_BF16_W_INT8"
31
+ },
32
+ {
33
+ "part": "single_pre",
34
+ "idx": 1,
35
+ "precision": "A_BF16_W_INT4"
36
+ },
37
+ {
38
+ "part": "single_post",
39
+ "idx": 1,
40
+ "precision": "A_BF16_W_INT4"
41
+ },
42
+ {
43
+ "part": "group_pre",
44
+ "idx": 2,
45
+ "precision": "A_BF16_W_INT8"
46
+ },
47
+ {
48
+ "part": "group_post",
49
+ "idx": 2,
50
+ "precision": "A_BF16_W_INT8"
51
+ },
52
+ {
53
+ "part": "single_pre",
54
+ "idx": 2,
55
+ "precision": "A_BF16_W_INT4"
56
+ },
57
+ {
58
+ "part": "single_post",
59
+ "idx": 2,
60
+ "precision": "A_BF16_W_INT4"
61
+ },
62
+ {
63
+ "part": "group_pre",
64
+ "idx": 3,
65
+ "precision": "A_BF16_W_INT8"
66
+ },
67
+ {
68
+ "part": "group_post",
69
+ "idx": 3,
70
+ "precision": "A_BF16_W_INT8"
71
+ },
72
+ {
73
+ "part": "single_pre",
74
+ "idx": 3,
75
+ "precision": "A_BF16_W_INT4"
76
+ },
77
+ {
78
+ "part": "single_post",
79
+ "idx": 3,
80
+ "precision": "A_BF16_W_INT4"
81
+ },
82
+ {
83
+ "part": "group_pre",
84
+ "idx": 4,
85
+ "precision": "A_BF16_W_INT8"
86
+ },
87
+ {
88
+ "part": "group_post",
89
+ "idx": 4,
90
+ "precision": "A_BF16_W_INT8"
91
+ },
92
+ {
93
+ "part": "single_pre",
94
+ "idx": 4,
95
+ "precision": "A_BF16_W_INT4"
96
+ },
97
+ {
98
+ "part": "single_post",
99
+ "idx": 4,
100
+ "precision": "A_BF16_W_INT4"
101
+ },
102
+ {
103
+ "part": "group_pre",
104
+ "idx": 5,
105
+ "precision": "A_BF16_W_INT8"
106
+ },
107
+ {
108
+ "part": "group_post",
109
+ "idx": 5,
110
+ "precision": "A_BF16_W_INT8"
111
+ },
112
+ {
113
+ "part": "single_pre",
114
+ "idx": 5,
115
+ "precision": "A_BF16_W_INT4"
116
+ },
117
+ {
118
+ "part": "single_post",
119
+ "idx": 5,
120
+ "precision": "A_BF16_W_INT4"
121
+ },
122
+ {
123
+ "part": "group_pre",
124
+ "idx": 6,
125
+ "precision": "A_BF16_W_INT8"
126
+ },
127
+ {
128
+ "part": "group_post",
129
+ "idx": 6,
130
+ "precision": "A_BF16_W_INT8"
131
+ },
132
+ {
133
+ "part": "single_pre",
134
+ "idx": 6,
135
+ "precision": "A_BF16_W_INT4"
136
+ },
137
+ {
138
+ "part": "single_post",
139
+ "idx": 6,
140
+ "precision": "A_BF16_W_INT4"
141
+ },
142
+ {
143
+ "part": "group_pre",
144
+ "idx": 7,
145
+ "precision": "A_BF16_W_INT8"
146
+ },
147
+ {
148
+ "part": "group_post",
149
+ "idx": 7,
150
+ "precision": "A_BF16_W_INT8"
151
+ },
152
+ {
153
+ "part": "single_pre",
154
+ "idx": 7,
155
+ "precision": "A_BF16_W_INT4"
156
+ },
157
+ {
158
+ "part": "single_post",
159
+ "idx": 7,
160
+ "precision": "A_BF16_W_INT4"
161
+ },
162
+ {
163
+ "part": "group_pre",
164
+ "idx": 8,
165
+ "precision": "A_BF16_W_INT8"
166
+ },
167
+ {
168
+ "part": "group_post",
169
+ "idx": 8,
170
+ "precision": "A_BF16_W_INT8"
171
+ },
172
+ {
173
+ "part": "single_pre",
174
+ "idx": 8,
175
+ "precision": "A_BF16_W_INT4"
176
+ },
177
+ {
178
+ "part": "single_post",
179
+ "idx": 8,
180
+ "precision": "A_BF16_W_INT4"
181
+ },
182
+ {
183
+ "part": "group_pre",
184
+ "idx": 9,
185
+ "precision": "A_BF16_W_INT8"
186
+ },
187
+ {
188
+ "part": "group_post",
189
+ "idx": 9,
190
+ "precision": "A_BF16_W_INT8"
191
+ },
192
+ {
193
+ "part": "single_pre",
194
+ "idx": 9,
195
+ "precision": "A_BF16_W_INT4"
196
+ },
197
+ {
198
+ "part": "single_post",
199
+ "idx": 9,
200
+ "precision": "A_BF16_W_INT4"
201
+ },
202
+ {
203
+ "part": "group_pre",
204
+ "idx": 10,
205
+ "precision": "A_BF16_W_INT8"
206
+ },
207
+ {
208
+ "part": "group_post",
209
+ "idx": 10,
210
+ "precision": "A_BF16_W_INT8"
211
+ },
212
+ {
213
+ "part": "single_pre",
214
+ "idx": 10,
215
+ "precision": "A_BF16_W_INT4"
216
+ },
217
+ {
218
+ "part": "single_post",
219
+ "idx": 10,
220
+ "precision": "A_BF16_W_INT4"
221
+ },
222
+ {
223
+ "part": "group_pre",
224
+ "idx": 11,
225
+ "precision": "A_BF16_W_INT8"
226
+ },
227
+ {
228
+ "part": "group_post",
229
+ "idx": 11,
230
+ "precision": "A_BF16_W_INT8"
231
+ },
232
+ {
233
+ "part": "single_pre",
234
+ "idx": 11,
235
+ "precision": "A_BF16_W_INT4"
236
+ },
237
+ {
238
+ "part": "single_post",
239
+ "idx": 11,
240
+ "precision": "A_BF16_W_INT4"
241
+ },
242
+ {
243
+ "part": "group_pre",
244
+ "idx": 12,
245
+ "precision": "A_BF16_W_INT8"
246
+ },
247
+ {
248
+ "part": "group_post",
249
+ "idx": 12,
250
+ "precision": "A_BF16_W_INT8"
251
+ },
252
+ {
253
+ "part": "single_pre",
254
+ "idx": 12,
255
+ "precision": "A_BF16_W_INT4"
256
+ },
257
+ {
258
+ "part": "single_post",
259
+ "idx": 12,
260
+ "precision": "A_BF16_W_INT4"
261
+ },
262
+ {
263
+ "part": "group_pre",
264
+ "idx": 13,
265
+ "precision": "A_BF16_W_INT8"
266
+ },
267
+ {
268
+ "part": "group_post",
269
+ "idx": 13,
270
+ "precision": "A_BF16_W_INT8"
271
+ },
272
+ {
273
+ "part": "single_pre",
274
+ "idx": 13,
275
+ "precision": "A_BF16_W_INT4"
276
+ },
277
+ {
278
+ "part": "single_post",
279
+ "idx": 13,
280
+ "precision": "A_BF16_W_INT4"
281
+ },
282
+ {
283
+ "part": "group_pre",
284
+ "idx": 14,
285
+ "precision": "A_BF16_W_INT8"
286
+ },
287
+ {
288
+ "part": "group_post",
289
+ "idx": 14,
290
+ "precision": "A_BF16_W_INT8"
291
+ },
292
+ {
293
+ "part": "single_pre",
294
+ "idx": 14,
295
+ "precision": "A_BF16_W_INT4"
296
+ },
297
+ {
298
+ "part": "single_post",
299
+ "idx": 14,
300
+ "precision": "A_BF16_W_INT4"
301
+ },
302
+ {
303
+ "part": "group_pre",
304
+ "idx": 15,
305
+ "precision": "A_BF16_W_INT8"
306
+ },
307
+ {
308
+ "part": "group_post",
309
+ "idx": 15,
310
+ "precision": "A_BF16_W_INT8"
311
+ },
312
+ {
313
+ "part": "single_pre",
314
+ "idx": 15,
315
+ "precision": "A_BF16_W_INT4"
316
+ },
317
+ {
318
+ "part": "single_post",
319
+ "idx": 15,
320
+ "precision": "A_BF16_W_INT4"
321
+ },
322
+ {
323
+ "part": "group_pre",
324
+ "idx": 16,
325
+ "precision": "A_BF16_W_INT8"
326
+ },
327
+ {
328
+ "part": "group_post",
329
+ "idx": 16,
330
+ "precision": "A_BF16_W_INT8"
331
+ },
332
+ {
333
+ "part": "single_pre",
334
+ "idx": 16,
335
+ "precision": "A_BF16_W_INT4"
336
+ },
337
+ {
338
+ "part": "single_post",
339
+ "idx": 16,
340
+ "precision": "A_BF16_W_INT4"
341
+ },
342
+ {
343
+ "part": "group_pre",
344
+ "idx": 17,
345
+ "precision": "A_BF16_W_INT8"
346
+ },
347
+ {
348
+ "part": "group_post",
349
+ "idx": 17,
350
+ "precision": "A_BF16_W_INT8"
351
+ },
352
+ {
353
+ "part": "single_pre",
354
+ "idx": 17,
355
+ "precision": "A_BF16_W_INT4"
356
+ },
357
+ {
358
+ "part": "single_post",
359
+ "idx": 17,
360
+ "precision": "A_BF16_W_INT4"
361
+ },
362
+ {
363
+ "part": "group_pre",
364
+ "idx": 18,
365
+ "precision": "A_BF16_W_INT8"
366
+ },
367
+ {
368
+ "part": "group_post",
369
+ "idx": 18,
370
+ "precision": "A_BF16_W_INT8"
371
+ },
372
+ {
373
+ "part": "single_pre",
374
+ "idx": 18,
375
+ "precision": "A_BF16_W_INT4"
376
+ },
377
+ {
378
+ "part": "single_post",
379
+ "idx": 18,
380
+ "precision": "A_BF16_W_INT4"
381
+ },
382
+ {
383
+ "part": "group_pre",
384
+ "idx": 19,
385
+ "precision": "A_BF16_W_INT8"
386
+ },
387
+ {
388
+ "part": "group_post",
389
+ "idx": 19,
390
+ "precision": "A_BF16_W_INT8"
391
+ },
392
+ {
393
+ "part": "single_pre",
394
+ "idx": 19,
395
+ "precision": "A_BF16_W_INT4"
396
+ },
397
+ {
398
+ "part": "single_post",
399
+ "idx": 19,
400
+ "precision": "A_BF16_W_INT4"
401
+ },
402
+ {
403
+ "part": "group_pre",
404
+ "idx": 20,
405
+ "precision": "A_BF16_W_INT8"
406
+ },
407
+ {
408
+ "part": "group_post",
409
+ "idx": 20,
410
+ "precision": "A_BF16_W_INT8"
411
+ },
412
+ {
413
+ "part": "single_pre",
414
+ "idx": 20,
415
+ "precision": "A_BF16_W_INT4"
416
+ },
417
+ {
418
+ "part": "single_post",
419
+ "idx": 20,
420
+ "precision": "A_BF16_W_INT4"
421
+ },
422
+ {
423
+ "part": "group_pre",
424
+ "idx": 21,
425
+ "precision": "A_BF16_W_INT8"
426
+ },
427
+ {
428
+ "part": "group_post",
429
+ "idx": 21,
430
+ "precision": "A_BF16_W_INT8"
431
+ },
432
+ {
433
+ "part": "single_pre",
434
+ "idx": 21,
435
+ "precision": "A_BF16_W_INT4"
436
+ },
437
+ {
438
+ "part": "single_post",
439
+ "idx": 21,
440
+ "precision": "A_BF16_W_INT4"
441
+ },
442
+ {
443
+ "part": "group_pre",
444
+ "idx": 22,
445
+ "precision": "A_BF16_W_INT8"
446
+ },
447
+ {
448
+ "part": "group_post",
449
+ "idx": 22,
450
+ "precision": "A_BF16_W_INT8"
451
+ },
452
+ {
453
+ "part": "single_pre",
454
+ "idx": 22,
455
+ "precision": "A_BF16_W_INT4"
456
+ },
457
+ {
458
+ "part": "single_post",
459
+ "idx": 22,
460
+ "precision": "A_BF16_W_INT4"
461
+ },
462
+ {
463
+ "part": "group_pre",
464
+ "idx": 23,
465
+ "precision": "A_BF16_W_INT8"
466
+ },
467
+ {
468
+ "part": "group_post",
469
+ "idx": 23,
470
+ "precision": "A_BF16_W_INT8"
471
+ },
472
+ {
473
+ "part": "single_pre",
474
+ "idx": 23,
475
+ "precision": "A_BF16_W_INT4"
476
+ },
477
+ {
478
+ "part": "single_post",
479
+ "idx": 23,
480
+ "precision": "A_BF16_W_INT4"
481
+ },
482
+ {
483
+ "part": "group_pre",
484
+ "idx": 24,
485
+ "precision": "A_BF16_W_INT8"
486
+ },
487
+ {
488
+ "part": "group_post",
489
+ "idx": 24,
490
+ "precision": "A_BF16_W_INT8"
491
+ },
492
+ {
493
+ "part": "single_pre",
494
+ "idx": 24,
495
+ "precision": "A_BF16_W_INT4"
496
+ },
497
+ {
498
+ "part": "single_post",
499
+ "idx": 24,
500
+ "precision": "A_BF16_W_INT4"
501
+ },
502
+ {
503
+ "part": "group_pre",
504
+ "idx": 25,
505
+ "precision": "A_BF16_W_INT8"
506
+ },
507
+ {
508
+ "part": "group_post",
509
+ "idx": 25,
510
+ "precision": "A_BF16_W_INT8"
511
+ },
512
+ {
513
+ "part": "single_pre",
514
+ "idx": 25,
515
+ "precision": "A_BF16_W_INT4"
516
+ },
517
+ {
518
+ "part": "single_post",
519
+ "idx": 25,
520
+ "precision": "A_BF16_W_INT4"
521
+ },
522
+ {
523
+ "part": "group_pre",
524
+ "idx": 26,
525
+ "precision": "A_BF16_W_INT8"
526
+ },
527
+ {
528
+ "part": "group_post",
529
+ "idx": 26,
530
+ "precision": "A_BF16_W_INT8"
531
+ },
532
+ {
533
+ "part": "single_pre",
534
+ "idx": 26,
535
+ "precision": "A_BF16_W_INT4"
536
+ },
537
+ {
538
+ "part": "single_post",
539
+ "idx": 26,
540
+ "precision": "A_BF16_W_INT4"
541
+ },
542
+ {
543
+ "part": "group_pre",
544
+ "idx": 27,
545
+ "precision": "A_BF16_W_INT8"
546
+ },
547
+ {
548
+ "part": "group_post",
549
+ "idx": 27,
550
+ "precision": "A_BF16_W_INT8"
551
+ },
552
+ {
553
+ "part": "single_pre",
554
+ "idx": 27,
555
+ "precision": "A_BF16_W_INT4"
556
+ },
557
+ {
558
+ "part": "single_post",
559
+ "idx": 27,
560
+ "precision": "A_BF16_W_INT4"
561
+ },
562
+ {
563
+ "part": "group_pre",
564
+ "idx": 28,
565
+ "precision": "A_BF16_W_INT8"
566
+ },
567
+ {
568
+ "part": "group_post",
569
+ "idx": 28,
570
+ "precision": "A_BF16_W_INT8"
571
+ },
572
+ {
573
+ "part": "single_pre",
574
+ "idx": 28,
575
+ "precision": "A_BF16_W_INT4"
576
+ },
577
+ {
578
+ "part": "single_post",
579
+ "idx": 28,
580
+ "precision": "A_BF16_W_INT4"
581
+ },
582
+ {
583
+ "part": "group_pre",
584
+ "idx": 29,
585
+ "precision": "A_BF16_W_INT8"
586
+ },
587
+ {
588
+ "part": "group_post",
589
+ "idx": 29,
590
+ "precision": "A_BF16_W_INT8"
591
+ },
592
+ {
593
+ "part": "single_pre",
594
+ "idx": 29,
595
+ "precision": "A_BF16_W_INT4"
596
+ },
597
+ {
598
+ "part": "single_post",
599
+ "idx": 29,
600
+ "precision": "A_BF16_W_INT4"
601
+ },
602
+ {
603
+ "part": "group_pre",
604
+ "idx": 30,
605
+ "precision": "A_BF16_W_INT8"
606
+ },
607
+ {
608
+ "part": "group_post",
609
+ "idx": 30,
610
+ "precision": "A_BF16_W_INT8"
611
+ },
612
+ {
613
+ "part": "single_pre",
614
+ "idx": 30,
615
+ "precision": "A_BF16_W_INT4"
616
+ },
617
+ {
618
+ "part": "single_post",
619
+ "idx": 30,
620
+ "precision": "A_BF16_W_INT4"
621
+ },
622
+ {
623
+ "part": "group_pre",
624
+ "idx": 31,
625
+ "precision": "A_BF16_W_INT8"
626
+ },
627
+ {
628
+ "part": "group_post",
629
+ "idx": 31,
630
+ "precision": "A_BF16_W_INT8"
631
+ },
632
+ {
633
+ "part": "single_pre",
634
+ "idx": 31,
635
+ "precision": "A_BF16_W_INT4"
636
+ },
637
+ {
638
+ "part": "single_post",
639
+ "idx": 31,
640
+ "precision": "A_BF16_W_INT4"
641
+ },
642
+ {
643
+ "part": "group_pre",
644
+ "idx": 32,
645
+ "precision": "A_BF16_W_INT8"
646
+ },
647
+ {
648
+ "part": "group_post",
649
+ "idx": 32,
650
+ "precision": "A_BF16_W_INT8"
651
+ },
652
+ {
653
+ "part": "single_pre",
654
+ "idx": 32,
655
+ "precision": "A_BF16_W_INT4"
656
+ },
657
+ {
658
+ "part": "single_post",
659
+ "idx": 32,
660
+ "precision": "A_BF16_W_INT4"
661
+ },
662
+ {
663
+ "part": "group_pre",
664
+ "idx": 33,
665
+ "precision": "A_BF16_W_INT8"
666
+ },
667
+ {
668
+ "part": "group_post",
669
+ "idx": 33,
670
+ "precision": "A_BF16_W_INT8"
671
+ },
672
+ {
673
+ "part": "single_pre",
674
+ "idx": 33,
675
+ "precision": "A_BF16_W_INT4"
676
+ },
677
+ {
678
+ "part": "single_post",
679
+ "idx": 33,
680
+ "precision": "A_BF16_W_INT4"
681
+ },
682
+ {
683
+ "part": "group_pre",
684
+ "idx": 34,
685
+ "precision": "A_BF16_W_INT8"
686
+ },
687
+ {
688
+ "part": "group_post",
689
+ "idx": 34,
690
+ "precision": "A_BF16_W_INT8"
691
+ },
692
+ {
693
+ "part": "single_pre",
694
+ "idx": 34,
695
+ "precision": "A_BF16_W_INT4"
696
+ },
697
+ {
698
+ "part": "single_post",
699
+ "idx": 34,
700
+ "precision": "A_BF16_W_INT4"
701
+ },
702
+ {
703
+ "part": "group_pre",
704
+ "idx": 35,
705
+ "precision": "A_BF16_W_INT8"
706
+ },
707
+ {
708
+ "part": "single_pre",
709
+ "idx": 35,
710
+ "precision": "A_BF16_W_INT4"
711
+ },
712
+ {
713
+ "part": "single_post",
714
+ "idx": 35,
715
+ "precision": "A_BF16_W_INT4"
716
+ },
717
+ {
718
+ "part": "group_cache",
719
+ "idx": 0,
720
+ "precision": "A_BF16_W_INT8"
721
+ },
722
+ {
723
+ "part": "group_cache",
724
+ "idx": 128,
725
+ "precision": "A_BF16_W_INT8"
726
+ },
727
+ {
728
+ "part": "group_cache",
729
+ "idx": 256,
730
+ "precision": "A_BF16_W_INT8"
731
+ },
732
+ {
733
+ "part": "group_cache",
734
+ "idx": 384,
735
+ "precision": "A_BF16_W_INT8"
736
+ },
737
+ {
738
+ "part": "group_cache",
739
+ "idx": 512,
740
+ "precision": "A_BF16_W_INT8"
741
+ },
742
+ {
743
+ "part": "group_cache",
744
+ "idx": 640,
745
+ "precision": "A_BF16_W_INT8"
746
+ },
747
+ {
748
+ "part": "group_cache",
749
+ "idx": 768,
750
+ "precision": "A_BF16_W_INT8"
751
+ },
752
+ {
753
+ "part": "group_cache",
754
+ "idx": 896,
755
+ "precision": "A_BF16_W_INT8"
756
+ },
757
+ {
758
+ "part": "group_cache",
759
+ "idx": 1024,
760
+ "precision": "A_BF16_W_INT8"
761
+ },
762
+ {
763
+ "part": "group_cache",
764
+ "idx": 1152,
765
+ "precision": "A_BF16_W_INT8"
766
+ },
767
+ {
768
+ "part": "group_cache",
769
+ "idx": 1280,
770
+ "precision": "A_BF16_W_INT8"
771
+ },
772
+ {
773
+ "part": "group_cache",
774
+ "idx": 1408,
775
+ "precision": "A_BF16_W_INT8"
776
+ },
777
+ {
778
+ "part": "group_cache",
779
+ "idx": 1536,
780
+ "precision": "A_BF16_W_INT8"
781
+ },
782
+ {
783
+ "part": "group_cache",
784
+ "idx": 1664,
785
+ "precision": "A_BF16_W_INT8"
786
+ },
787
+ {
788
+ "part": "group_cache",
789
+ "idx": 1792,
790
+ "precision": "A_BF16_W_INT8"
791
+ },
792
+ {
793
+ "part": "group_cache",
794
+ "idx": 1920,
795
+ "precision": "A_BF16_W_INT8"
796
+ },
797
+ {
798
+ "part": "single_cache",
799
+ "idx": 127,
800
+ "precision": "A_BF16_W_INT4"
801
+ },
802
+ {
803
+ "part": "single_cache",
804
+ "idx": 255,
805
+ "precision": "A_BF16_W_INT4"
806
+ },
807
+ {
808
+ "part": "single_cache",
809
+ "idx": 383,
810
+ "precision": "A_BF16_W_INT4"
811
+ },
812
+ {
813
+ "part": "single_cache",
814
+ "idx": 511,
815
+ "precision": "A_BF16_W_INT4"
816
+ },
817
+ {
818
+ "part": "single_cache",
819
+ "idx": 639,
820
+ "precision": "A_BF16_W_INT4"
821
+ },
822
+ {
823
+ "part": "single_cache",
824
+ "idx": 767,
825
+ "precision": "A_BF16_W_INT4"
826
+ },
827
+ {
828
+ "part": "single_cache",
829
+ "idx": 895,
830
+ "precision": "A_BF16_W_INT4"
831
+ },
832
+ {
833
+ "part": "single_cache",
834
+ "idx": 1023,
835
+ "precision": "A_BF16_W_INT4"
836
+ },
837
+ {
838
+ "part": "single_cache",
839
+ "idx": 1151,
840
+ "precision": "A_BF16_W_INT4"
841
+ },
842
+ {
843
+ "part": "single_cache",
844
+ "idx": 1279,
845
+ "precision": "A_BF16_W_INT4"
846
+ },
847
+ {
848
+ "part": "single_cache",
849
+ "idx": 1407,
850
+ "precision": "A_BF16_W_INT4"
851
+ },
852
+ {
853
+ "part": "single_cache",
854
+ "idx": 1535,
855
+ "precision": "A_BF16_W_INT4"
856
+ },
857
+ {
858
+ "part": "single_cache",
859
+ "idx": 1663,
860
+ "precision": "A_BF16_W_INT4"
861
+ },
862
+ {
863
+ "part": "single_cache",
864
+ "idx": 1791,
865
+ "precision": "A_BF16_W_INT4"
866
+ },
867
+ {
868
+ "part": "single_cache",
869
+ "idx": 1919,
870
+ "precision": "A_BF16_W_INT4"
871
+ },
872
+ {
873
+ "part": "single_cache",
874
+ "idx": 2047,
875
+ "precision": "A_BF16_W_INT4"
876
+ },
877
+ {
878
+ "part": "vision",
879
+ "idx": 0,
880
+ "precision": "A_BF16_W_INT8"
881
+ }
882
+ ]
devkit/preprocessor_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "size": {
3
+ "longest_edge": 16777216,
4
+ "shortest_edge": 65536
5
+ },
6
+ "patch_size": 16,
7
+ "temporal_patch_size": 2,
8
+ "merge_size": 2,
9
+ "image_mean": [
10
+ 0.5,
11
+ 0.5,
12
+ 0.5
13
+ ],
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "processor_class": "Qwen3VLProcessor",
20
+ "image_processor_type": "Qwen2VLImageProcessorFast"
21
+ }
devkit/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
devkit/tokenizer_config.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|im_start|>",
216
+ "<|im_end|>",
217
+ "<|object_ref_start|>",
218
+ "<|object_ref_end|>",
219
+ "<|box_start|>",
220
+ "<|box_end|>",
221
+ "<|quad_start|>",
222
+ "<|quad_end|>",
223
+ "<|vision_start|>",
224
+ "<|vision_end|>",
225
+ "<|vision_pad|>",
226
+ "<|image_pad|>",
227
+ "<|video_pad|>"
228
+ ],
229
+ "bos_token": null,
230
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0].role == 'system' %}\n {%- if messages[0].content is string %}\n {{- messages[0].content }}\n {%- else %}\n {%- for content in messages[0].content %}\n {%- if 'text' in content %}\n {{- content.text }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {{- '\\n\\n' }}\n {%- endif %}\n {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0].content is string %}\n {{- messages[0].content }}\n {%- else %}\n {%- for content in messages[0].content %}\n {%- if 'text' in content %}\n {{- content.text }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- set image_count = namespace(value=0) %}\n{%- set video_count = namespace(value=0) %}\n{%- for message in messages %}\n {%- if message.role == \"user\" %}\n {{- '<|im_start|>' + message.role + '\\n' }}\n {%- if message.content is string %}\n {{- message.content }}\n {%- else %}\n {%- for content in message.content %}\n {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}\n {%- set image_count.value = image_count.value + 1 %}\n {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}\n <|vision_start|><|image_pad|><|vision_end|>\n {%- elif content.type == 'video' or 'video' in content %}\n {%- set video_count.value = video_count.value + 1 %}\n {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}\n <|vision_start|><|video_pad|><|vision_end|>\n {%- elif 'text' in content %}\n {{- content.text }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role + '\\n' }}\n {%- if message.content is string %}\n {{- message.content }}\n {%- else %}\n {%- for content_item in message.content %}\n {%- if 'text' in content_item %}\n {{- content_item.text }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {%- if message.tool_calls %}\n {%- for tool_call in message.tool_calls %}\n {%- if (loop.first and message.content) or (not loop.first) %}\n {{- '\\n' }}\n {%- endif %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {%- if tool_call.arguments is string %}\n {{- tool_call.arguments }}\n {%- else %}\n {{- tool_call.arguments | tojson }}\n {%- endif %}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {%- if message.content is string %}\n {{- message.content }}\n {%- else %}\n {%- for content in message.content %}\n {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}\n {%- set image_count.value = image_count.value + 1 %}\n {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}\n <|vision_start|><|image_pad|><|vision_end|>\n {%- elif content.type == 'video' or 'video' in content %}\n {%- set video_count.value = video_count.value + 1 %}\n {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}\n <|vision_start|><|video_pad|><|vision_end|>\n {%- elif 'text' in content %}\n {{- content.text }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
231
+ "clean_up_tokenization_spaces": false,
232
+ "eos_token": "<|im_end|>",
233
+ "errors": "replace",
234
+ "model_max_length": 262144,
235
+ "pad_token": "<|endoftext|>",
236
+ "split_special_tokens": false,
237
+ "tokenizer_class": "Qwen2Tokenizer",
238
+ "unk_token": null
239
+ }
devkit/video_preprocessor_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "size": {
3
+ "longest_edge": 25165824,
4
+ "shortest_edge": 4096
5
+ },
6
+ "patch_size": 16,
7
+ "temporal_patch_size": 2,
8
+ "merge_size": 2,
9
+ "image_mean": [
10
+ 0.5,
11
+ 0.5,
12
+ 0.5
13
+ ],
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "processor_class": "Qwen3VLProcessor",
20
+ "video_processor_type": "Qwen3VLVideoProcessor"
21
+ }
devkit/vlm_config.json ADDED
@@ -0,0 +1,159 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_name": "Qwen3-VL-4B-Instruct",
3
+ "model_type": "vlm-qwen3_vl",
4
+ "vm_cfg": {
5
+ "model_type": "qwen3_vl",
6
+ "arch": "qwen3_vl",
7
+ "image_size": [
8
+ 256,
9
+ 256
10
+ ],
11
+ "patch_size": 16,
12
+ "cls_embed": false,
13
+ "hidden_size": 1024,
14
+ "intermediate_size": 4096,
15
+ "num_attention_heads": 16,
16
+ "num_hidden_layers": 24,
17
+ "hidden_act": "gelu_pytorch_tanh",
18
+ "layer_norm_eps": 1e-06,
19
+ "spatial_merge_size": 2,
20
+ "temporal_patch_size": 2,
21
+ "window_size": 0,
22
+ "num_position_embeddings": 2304,
23
+ "fullatt_block_indexes": [],
24
+ "deepstack_visual_indexes": [
25
+ 5,
26
+ 11,
27
+ 17
28
+ ]
29
+ },
30
+ "mm_cfg": {
31
+ "num_layers": 2,
32
+ "hidden_act": "gelu",
33
+ "mm_tokens_per_image": 64,
34
+ "proj_dim": 2560,
35
+ "downsample_factor": 1,
36
+ "projector_use_layernorm": true
37
+ },
38
+ "lm_cfg": {
39
+ "model_type": "qwen3_vl_text",
40
+ "data_type": "bfloat16",
41
+ "arch": "qwen",
42
+ "token_cfg": {
43
+ "vocab_size": 151936
44
+ },
45
+ "rope_cfg": {
46
+ "rope_theta": 5000000,
47
+ "rope_local_base_freq": 10000,
48
+ "rope_scaling": {
49
+ "factor": 1.0,
50
+ "low_freq_factor": 0,
51
+ "high_freq_factor": 0,
52
+ "original_max_position_embeddings": 0,
53
+ "long_factor": null,
54
+ "short_factor": null,
55
+ "rope_type": "default",
56
+ "mrope_section": null,
57
+ "mrope_interleaved": false
58
+ }
59
+ },
60
+ "attn_cfg": {
61
+ "num_attention_heads": 32,
62
+ "num_key_value_heads": 8,
63
+ "head_dim": 128,
64
+ "swa_enable": false,
65
+ "sliding_window": 0,
66
+ "attention_bias": false,
67
+ "attention_dropout": 0.0,
68
+ "query_pre_attn_scalar": 0
69
+ },
70
+ "mlp_cfg": {
71
+ "intermediate_size": 9728,
72
+ "act": "silu",
73
+ "num_layers": 3,
74
+ "mlp_bias": false
75
+ },
76
+ "hidden_size": 2560,
77
+ "num_hidden_layers": 36,
78
+ "max_position_embeddings": 2048,
79
+ "rms_norm_eps": 1e-06,
80
+ "rms_norm_unit_offset": false,
81
+ "layer_types": [
82
+ "full_attention",
83
+ "full_attention",
84
+ "full_attention",
85
+ "full_attention",
86
+ "full_attention",
87
+ "full_attention",
88
+ "full_attention",
89
+ "full_attention",
90
+ "full_attention",
91
+ "full_attention",
92
+ "full_attention",
93
+ "full_attention",
94
+ "full_attention",
95
+ "full_attention",
96
+ "full_attention",
97
+ "full_attention",
98
+ "full_attention",
99
+ "full_attention",
100
+ "full_attention",
101
+ "full_attention",
102
+ "full_attention",
103
+ "full_attention",
104
+ "full_attention",
105
+ "full_attention",
106
+ "full_attention",
107
+ "full_attention",
108
+ "full_attention",
109
+ "full_attention",
110
+ "full_attention",
111
+ "full_attention",
112
+ "full_attention",
113
+ "full_attention",
114
+ "full_attention",
115
+ "full_attention",
116
+ "full_attention",
117
+ "full_attention"
118
+ ],
119
+ "attn_logit_softcapping": null,
120
+ "final_logit_softcapping": null,
121
+ "lm_head_num_splits": 3,
122
+ "lm_head_split_dim": 50656,
123
+ "conv_L_cache": 3,
124
+ "conv_bias": false,
125
+ "lora_cfg": null
126
+ },
127
+ "pipeline_cfg": {
128
+ "system_prompt": null,
129
+ "chat_template": null,
130
+ "max_num_tokens": 2048,
131
+ "input_token_group_size": 128,
132
+ "input_token_group_offsets": [
133
+ 0,
134
+ 128,
135
+ 256,
136
+ 384,
137
+ 512,
138
+ 640,
139
+ 768,
140
+ 896,
141
+ 1024,
142
+ 1152,
143
+ 1280,
144
+ 1408,
145
+ 1536,
146
+ 1664,
147
+ 1792,
148
+ 1920
149
+ ],
150
+ "future_token_mask_size": 128,
151
+ "return_logits": false,
152
+ "use_strided_kv_cache": false,
153
+ "enable_filter_sharing": false,
154
+ "quantize_embeddings": false,
155
+ "split_mlp": true
156
+ },
157
+ "language_model_name": "Qwen3-VL-4B-Instruct_language",
158
+ "vision_model_name": "Qwen3-VL-4B-Instruct_vision"
159
+ }
devkit/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token0_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4fbdaca51ac929f6fc85d1ae0d725983920f909eb39088978cc812521485f13
3
+ size 4249408
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1024_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bcf507b39784a04c44a41f9b5d40e73c53e0dc7296e02a639b5f43f29e4b8ba
3
+ size 11966104
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1152_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32ffc47e9ca8966c8c51b5cd3435bedbd3f445769e9dbb510f4932f6c4027fb3
3
+ size 14331336
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1280_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:362621ddd71a108881722bc70fbb9b4338eb26a26e0c3c84b3823a1d920200d9
3
+ size 13895896
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token128_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6ec9010f2d3c9f664f724c8416405dbe42f1e897451f6c844f63c446ae24594
3
+ size 4599040
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1408_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d4a934d50ea014a0757287cf2dcbd681f2fe127938683d05c30e0d49d50bb09
3
+ size 16270480
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1536_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca4bc39c9914865ec18c8c72d4cd5836adc4de4e4e913ed988d8193be46b4d78
3
+ size 17263480
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1664_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3c190bbae478f1362a5313793098f02839ced342d4c4f7d835c2579b4de9d5e9
3
+ size 17058072
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1792_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5cc2f1a5b000480edc1cb6be64482e9c96f33e6a4a490da84bf6f447f79ed385
3
+ size 17340936
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token1920_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b92247911e6e18e62323cdeb4a4868b0781f823cc0eaa6c891a5c80e070003bb
3
+ size 18246864
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token256_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:443b1148ca7141cf8eb0d31f4161227d9ba87f84c60bf2091d16be2445a674e0
3
+ size 4725968
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token384_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c159d3b887d9532e4b09bb80b018c404c7959a5c8450b6338aed793d3ce453c
3
+ size 6137200
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token512_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e5511f228ec4c4b10381a0b4e6177079310029d9dcebd13d7dc1d821cc56eaa
3
+ size 6502112
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token640_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb01bce84b95571928042c76f8ee367eebb1eb5318c10998f2fa7e9abed4a77d
3
+ size 8004480
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token768_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99f8e3c9dd2bbd3bf91ea5b773bcce78227a5889a346dfe4144c59bdceec0aae
3
+ size 10267480
elf_files/Qwen3-VL-4B-Instruct_language_n128_cache_token896_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf83385734d94ebe1a761d3a8f61dfe91617003f876bb4e4258bc584189639e6
3
+ size 10611240
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer0_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:def50360725a97edc23c2ddf0a0f0dc31762b5bd6c4fec4d447cad35c5c11436
3
+ size 94606128
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer10_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da631c4523e061de2987cea6af16cdb1802c3ae9626b8ed6672f3b84fe7739db
3
+ size 94001456
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer11_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a8e9f98006af381330e44df0fe4369ee2f1c7d0e4dd61f356cbd74998c8f174
3
+ size 94090256
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer12_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2716d404123cd39dd6e8e66235e20b3579a187629eaf485ee7e2fb0c7b0acc51
3
+ size 93791840
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer13_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:871d2d9c1fa5d243410c45cfb8279f9be32d207d409e0e05e682ace5736dd9dc
3
+ size 93708144
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer14_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c7f5051159c614a1504ff47adc2b582280d796c4b2364f708eaecc5fa78d66e
3
+ size 93648048
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer15_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8499116822e3f12acb50f1ea60608810ceebed6812fde405427f5457c9f92b00
3
+ size 93587832
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer16_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47057fca00a0319eea3887d0d0e3c4877451ef4be2ca300096cc1327286ff45d
3
+ size 93681152
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer17_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55a5f18bbfd4b6d8739f974798aa1112c09573a3b5d772c3a7c9953a4164c690
3
+ size 93542336
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer18_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c3f878fa7d72a4999980f0f135955e667cfe1ecb9b32b12d9c30feec426c02d
3
+ size 93359776
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer19_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec929fba149070470e5e5dc25805a3e43b37881585e3e13f2eb7b92bc98a99f3
3
+ size 93122952
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer1_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41c15455e3c0356686595b8fa1da591d02cb4843a269d87c21a2cee37845d7a4
3
+ size 89518504
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer20_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f2d68ed2a15ea5fe146db8c3054e5199e3206b51c7d6732d001a6e903dba49c
3
+ size 93403192
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer21_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1fdea7e90bcf60764525a3bae24c363d7336769917e8bb7e7d26b7ef34b00ca
3
+ size 93299280
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer22_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0286113f128c62dd8b3e3e62af0cb918de5d5460203239cffb9af658979e6993
3
+ size 93170712
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer23_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bdcb84cf50089a70d27b48ecdbb0fa4f2931e516eb13050952bcc06ad3f19094
3
+ size 93443464
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer24_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb7e96ac2fc0ee43dfd5784a5d690ae2be3112300f93bdbf9accf13a7c7bb631
3
+ size 93577216
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer25_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:346d0cd2eef47befb2328baf810e7d45a406a4a66e925dda083bebf85fe72fd9
3
+ size 94031512
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer26_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac35fecd30118b562e841a55d3c7e9c2b16aeea1ed479665d2ff28b31549f959
3
+ size 94159096
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer27_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b57b2e71c4c7ccb461801df38b4d1df6a471be504b11e562569424993111095
3
+ size 94228960
elf_files/Qwen3-VL-4B-Instruct_language_n128_post_layer28_stage1_mla.elf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a02ea1b547f706265e22690891644fcdeeddfbdfb0711337e4f7fe4aeb2b21f
3
+ size 94238920