Hugging Face's logo

BCCard
/
Qwen2.5-VL-72B-Instruct-FP8-Dynamic

Image-to-Text
Transformers
Safetensors
English
qwen2_5_vl
vllm
vision
fp8
text-generation-inference
compressed-tensors
Model card Files
xet
Community
sh2orc commited on
Commit
be82053
·
verified ·
1 Parent(s): a3750e8

Upload folder using huggingface_hub

Browse files
Files changed (19) hide show
  1. config.json +1 -163
  2. model-00001-of-00016.safetensors +2 -2
  3. model-00002-of-00016.safetensors +2 -2
  4. model-00003-of-00016.safetensors +2 -2
  5. model-00004-of-00016.safetensors +2 -2
  6. model-00005-of-00016.safetensors +1 -1
  7. model-00006-of-00016.safetensors +1 -1
  8. model-00007-of-00016.safetensors +1 -1
  9. model-00008-of-00016.safetensors +1 -1
  10. model-00009-of-00016.safetensors +1 -1
  11. model-00010-of-00016.safetensors +1 -1
  12. model-00011-of-00016.safetensors +1 -1
  13. model-00012-of-00016.safetensors +1 -1
  14. model-00013-of-00016.safetensors +1 -1
  15. model-00014-of-00016.safetensors +1 -1
  16. model-00015-of-00016.safetensors +1 -1
  17. model-00016-of-00016.safetensors +2 -2
  18. model.safetensors.index.json +0 -0
  19. recipe.yaml +1 -1
config.json CHANGED
@@ -50,170 +50,8 @@
50
  }
51
  },
52
  "format": "float-quantized",
53
- "global_compression_ratio": null,
54
  "ignore": [
55
- "visual.blocks.0.attn.qkv",
56
- "visual.blocks.0.attn.proj",
57
- "visual.blocks.0.mlp.gate_proj",
58
- "visual.blocks.0.mlp.up_proj",
59
- "visual.blocks.0.mlp.down_proj",
60
- "visual.blocks.1.attn.qkv",
61
- "visual.blocks.1.attn.proj",
62
- "visual.blocks.1.mlp.gate_proj",
63
- "visual.blocks.1.mlp.up_proj",
64
- "visual.blocks.1.mlp.down_proj",
65
- "visual.blocks.2.attn.qkv",
66
- "visual.blocks.2.attn.proj",
67
- "visual.blocks.2.mlp.gate_proj",
68
- "visual.blocks.2.mlp.up_proj",
69
- "visual.blocks.2.mlp.down_proj",
70
- "visual.blocks.3.attn.qkv",
71
- "visual.blocks.3.attn.proj",
72
- "visual.blocks.3.mlp.gate_proj",
73
- "visual.blocks.3.mlp.up_proj",
74
- "visual.blocks.3.mlp.down_proj",
75
- "visual.blocks.4.attn.qkv",
76
- "visual.blocks.4.attn.proj",
77
- "visual.blocks.4.mlp.gate_proj",
78
- "visual.blocks.4.mlp.up_proj",
79
- "visual.blocks.4.mlp.down_proj",
80
- "visual.blocks.5.attn.qkv",
81
- "visual.blocks.5.attn.proj",
82
- "visual.blocks.5.mlp.gate_proj",
83
- "visual.blocks.5.mlp.up_proj",
84
- "visual.blocks.5.mlp.down_proj",
85
- "visual.blocks.6.attn.qkv",
86
- "visual.blocks.6.attn.proj",
87
- "visual.blocks.6.mlp.gate_proj",
88
- "visual.blocks.6.mlp.up_proj",
89
- "visual.blocks.6.mlp.down_proj",
90
- "visual.blocks.7.attn.qkv",
91
- "visual.blocks.7.attn.proj",
92
- "visual.blocks.7.mlp.gate_proj",
93
- "visual.blocks.7.mlp.up_proj",
94
- "visual.blocks.7.mlp.down_proj",
95
- "visual.blocks.8.attn.qkv",
96
- "visual.blocks.8.attn.proj",
97
- "visual.blocks.8.mlp.gate_proj",
98
- "visual.blocks.8.mlp.up_proj",
99
- "visual.blocks.8.mlp.down_proj",
100
- "visual.blocks.9.attn.qkv",
101
- "visual.blocks.9.attn.proj",
102
- "visual.blocks.9.mlp.gate_proj",
103
- "visual.blocks.9.mlp.up_proj",
104
- "visual.blocks.9.mlp.down_proj",
105
- "visual.blocks.10.attn.qkv",
106
- "visual.blocks.10.attn.proj",
107
- "visual.blocks.10.mlp.gate_proj",
108
- "visual.blocks.10.mlp.up_proj",
109
- "visual.blocks.10.mlp.down_proj",
110
- "visual.blocks.11.attn.qkv",
111
- "visual.blocks.11.attn.proj",
112
- "visual.blocks.11.mlp.gate_proj",
113
- "visual.blocks.11.mlp.up_proj",
114
- "visual.blocks.11.mlp.down_proj",
115
- "visual.blocks.12.attn.qkv",
116
- "visual.blocks.12.attn.proj",
117
- "visual.blocks.12.mlp.gate_proj",
118
- "visual.blocks.12.mlp.up_proj",
119
- "visual.blocks.12.mlp.down_proj",
120
- "visual.blocks.13.attn.qkv",
121
- "visual.blocks.13.attn.proj",
122
- "visual.blocks.13.mlp.gate_proj",
123
- "visual.blocks.13.mlp.up_proj",
124
- "visual.blocks.13.mlp.down_proj",
125
- "visual.blocks.14.attn.qkv",
126
- "visual.blocks.14.attn.proj",
127
- "visual.blocks.14.mlp.gate_proj",
128
- "visual.blocks.14.mlp.up_proj",
129
- "visual.blocks.14.mlp.down_proj",
130
- "visual.blocks.15.attn.qkv",
131
- "visual.blocks.15.attn.proj",
132
- "visual.blocks.15.mlp.gate_proj",
133
- "visual.blocks.15.mlp.up_proj",
134
- "visual.blocks.15.mlp.down_proj",
135
- "visual.blocks.16.attn.qkv",
136
- "visual.blocks.16.attn.proj",
137
- "visual.blocks.16.mlp.gate_proj",
138
- "visual.blocks.16.mlp.up_proj",
139
- "visual.blocks.16.mlp.down_proj",
140
- "visual.blocks.17.attn.qkv",
141
- "visual.blocks.17.attn.proj",
142
- "visual.blocks.17.mlp.gate_proj",
143
- "visual.blocks.17.mlp.up_proj",
144
- "visual.blocks.17.mlp.down_proj",
145
- "visual.blocks.18.attn.qkv",
146
- "visual.blocks.18.attn.proj",
147
- "visual.blocks.18.mlp.gate_proj",
148
- "visual.blocks.18.mlp.up_proj",
149
- "visual.blocks.18.mlp.down_proj",
150
- "visual.blocks.19.attn.qkv",
151
- "visual.blocks.19.attn.proj",
152
- "visual.blocks.19.mlp.gate_proj",
153
- "visual.blocks.19.mlp.up_proj",
154
- "visual.blocks.19.mlp.down_proj",
155
- "visual.blocks.20.attn.qkv",
156
- "visual.blocks.20.attn.proj",
157
- "visual.blocks.20.mlp.gate_proj",
158
- "visual.blocks.20.mlp.up_proj",
159
- "visual.blocks.20.mlp.down_proj",
160
- "visual.blocks.21.attn.qkv",
161
- "visual.blocks.21.attn.proj",
162
- "visual.blocks.21.mlp.gate_proj",
163
- "visual.blocks.21.mlp.up_proj",
164
- "visual.blocks.21.mlp.down_proj",
165
- "visual.blocks.22.attn.qkv",
166
- "visual.blocks.22.attn.proj",
167
- "visual.blocks.22.mlp.gate_proj",
168
- "visual.blocks.22.mlp.up_proj",
169
- "visual.blocks.22.mlp.down_proj",
170
- "visual.blocks.23.attn.qkv",
171
- "visual.blocks.23.attn.proj",
172
- "visual.blocks.23.mlp.gate_proj",
173
- "visual.blocks.23.mlp.up_proj",
174
- "visual.blocks.23.mlp.down_proj",
175
- "visual.blocks.24.attn.qkv",
176
- "visual.blocks.24.attn.proj",
177
- "visual.blocks.24.mlp.gate_proj",
178
- "visual.blocks.24.mlp.up_proj",
179
- "visual.blocks.24.mlp.down_proj",
180
- "visual.blocks.25.attn.qkv",
181
- "visual.blocks.25.attn.proj",
182
- "visual.blocks.25.mlp.gate_proj",
183
- "visual.blocks.25.mlp.up_proj",
184
- "visual.blocks.25.mlp.down_proj",
185
- "visual.blocks.26.attn.qkv",
186
- "visual.blocks.26.attn.proj",
187
- "visual.blocks.26.mlp.gate_proj",
188
- "visual.blocks.26.mlp.up_proj",
189
- "visual.blocks.26.mlp.down_proj",
190
- "visual.blocks.27.attn.qkv",
191
- "visual.blocks.27.attn.proj",
192
- "visual.blocks.27.mlp.gate_proj",
193
- "visual.blocks.27.mlp.up_proj",
194
- "visual.blocks.27.mlp.down_proj",
195
- "visual.blocks.28.attn.qkv",
196
- "visual.blocks.28.attn.proj",
197
- "visual.blocks.28.mlp.gate_proj",
198
- "visual.blocks.28.mlp.up_proj",
199
- "visual.blocks.28.mlp.down_proj",
200
- "visual.blocks.29.attn.qkv",
201
- "visual.blocks.29.attn.proj",
202
- "visual.blocks.29.mlp.gate_proj",
203
- "visual.blocks.29.mlp.up_proj",
204
- "visual.blocks.29.mlp.down_proj",
205
- "visual.blocks.30.attn.qkv",
206
- "visual.blocks.30.attn.proj",
207
- "visual.blocks.30.mlp.gate_proj",
208
- "visual.blocks.30.mlp.up_proj",
209
- "visual.blocks.30.mlp.down_proj",
210
- "visual.blocks.31.attn.qkv",
211
- "visual.blocks.31.attn.proj",
212
- "visual.blocks.31.mlp.gate_proj",
213
- "visual.blocks.31.mlp.up_proj",
214
- "visual.blocks.31.mlp.down_proj",
215
- "visual.merger.mlp.0",
216
- "visual.merger.mlp.2",
217
  "lm_head"
218
  ],
219
  "kv_cache_scheme": null,
 
50
  }
51
  },
52
  "format": "float-quantized",
53
+ "global_compression_ratio": 1.459839236827677,
54
  "ignore": [
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
55
  "lm_head"
56
  ],
57
  "kv_cache_scheme": null,
model-00001-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c4fdcf55d00ffa11e55fcc79ab5cca51d50a1d267219820ac66c36be0177fb23
3
- size 4929545176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0307c8cc9b66900a02879d8992cdd378ae55555cd0c7470555766e236922fbe
3
+ size 4954728712
model-00002-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:66dcd62e9496e9b8c456f6c95df848034fe648f4414ff1088a35f7a6efa44156
3
- size 4873985928
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58df71e437de60f9c8519d02006677ea7b5c51698f36c71402e7fba4f7beb3b0
3
+ size 4782759032
model-00003-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4f76e418f06f6b9b7f8a524f6405b1baee70ca7418a9a660f7537c490263ae35
3
- size 4782749304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2194ad0d8ee67df573cb1ad236c1154b45c67a700a76f496012e329d6109d751
3
+ size 4957904264
model-00004-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:09782507221803a2287fcacb4ef166317fee2e62839ad43caa4c2f9ec461f77c
3
- size 4873986032
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:64681b14de8b28842cf179e0c9eb3e9cbaa1b9fb3db330a752b66e6b49a7d5d9
3
+ size 4941111512
model-00005-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ab1e3dfb60f12733be7aae9274996f14814cc4680a707e1a201a0b627b4373ea
3
  size 4782749368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df02dda70aaf195df21b216ba788912d3516009d62b25e163e1bb19a04a131e1
3
  size 4782749368
model-00006-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:39f8a745a63bdd2552d0aaa25728108ccd66718611bb8f1b1d1998a86e5054a1
3
  size 4873986032
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da7a1eaef3feb2fca165d422ad3eb5e7b4a573ec4e95660ea83b68750214e8a0
3
  size 4873986032
model-00007-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2f452321174378d32090e7c94cffa3b43a4539e2f407f936b12a35639122005a
3
  size 4782749368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5900c3e1d6bd04b24b47c00898df7b9c43aa798d5fd319a9ef93bb300af80978
3
  size 4782749368
model-00008-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:87afe66c34acb2c5a7b118db74939afe8f47613f442eca4ed0f38a02e4e93fed
3
  size 4873986032
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef7a80c573b289a670291327f1ab7f5c5b4e1b6fcad2ae787c432865b085cef0
3
  size 4873986032
model-00009-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4a79addd10e3e66dcc4b53822ed86a7e31197ca3a54eaacc2a8140921d158f47
3
  size 4782749368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41f0524b3c09d411c9917925f68da75383ebcc361237dc8a87f9c855dc64cc85
3
  size 4782749368
model-00010-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:83154d2dd5e7064947716c0930013e17943c378f80343c5f6067e8afdc291c1f
3
  size 4873986032
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:345c23ba54449827f380b4fd3564393ace34d7d94c653d077554e4e24113260d
3
  size 4873986032
model-00011-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:48dd3dd98f76920c997a634e4f390f4ab11e629c7d76d752ef94844477bea183
3
  size 4782749368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b281a1c31b01bc2f011097f9c4aa8c73a07506cfb6f9f01059481ca0df083016
3
  size 4782749368
model-00012-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3168a6cbc214a1e83a04235e676fad12952ef36010442e02e007701ea23037e4
3
  size 4873986032
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d2cd32beedeab2c7db93560c0229d0bf32ee3caf3c1760f84ae1612d9f2edd9
3
  size 4873986032
model-00013-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:98d1b9f92f0154fc8f7cbfe0c6d7d3955b088cf2c520e1a41c2f157c07c5b6bf
3
  size 4782749368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0de062860e7e38dab7951e69c458edeb966ceea594edb25a310938be081d17a0
3
  size 4782749368
model-00014-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:46d25976bd3a6cb3055a1376a5a3ffec0f350d6a34a9749cbdd0495409e5e744
3
  size 4873986032
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:520389ce397216787010c5b7e133516d99fff04a3842121c8a786406fe215dc4
3
  size 4873986032
model-00015-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d17169ad645848ed1c32ef9d34529d80bbe052b3f5309d9f7973ed0fde30c09a
3
  size 4782749368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1151c6b0543c5dd4aa82c0235cadf01ed5e0450fbe4c1e37cc8c9a026e4c4e6e
3
  size 4782749368
model-00016-of-00016.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f3611e6a6ce31c332c9fcbaa2d5975ad981a6aa05470b66df74dab8811f7cb0a
3
- size 4096149840
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d39bbae5388ff0f6b2034be936a60cb726b16a5d9069811ab75eef8d6e0c0bf8
3
+ size 3218264712
model.safetensors.index.json CHANGED
The diff for this file is too large to render. See raw diff
 
recipe.yaml CHANGED
@@ -1,6 +1,6 @@
1
  default_stage:
2
  default_modifiers:
3
  QuantizationModifier:
4
- ignore: [lm_head, 're:visual.*']
5
  targets: [Linear]
6
  scheme: FP8_DYNAMIC
 
1
  default_stage:
2
  default_modifiers:
3
  QuantizationModifier:
4
+ ignore: ['re:.*lm_head', 're:vision_tower.*', 're:multi_modal_projector.*']
5
  targets: [Linear]
6
  scheme: FP8_DYNAMIC