Hugging Face's logo

helenai
/
Phi-3.5-vision-instruct-ov-fp16

OpenVINO
phi3_v
custom_code
Model card Files Community
helenai commited on
Commit
3e2b86c
·
1 Parent(s): 1c7ac95

Add model

Browse files
Files changed (19) hide show
  1. chat_template.json +3 -0
  2. config.json +0 -0
  3. generation_config.json +7 -0
  4. openvino_detokenizer.bin +3 -0
  5. openvino_detokenizer.xml +416 -0
  6. openvino_language_model.bin +3 -0
  7. openvino_language_model.xml +0 -0
  8. openvino_text_embeddings_model.bin +3 -0
  9. openvino_text_embeddings_model.xml +113 -0
  10. openvino_tokenizer.bin +3 -0
  11. openvino_tokenizer.xml +835 -0
  12. openvino_vision_embeddings_model.bin +3 -0
  13. openvino_vision_embeddings_model.xml +0 -0
  14. openvino_vision_projection_model.bin +3 -0
  15. openvino_vision_projection_model.xml +270 -0
  16. preprocessor_config.json +21 -0
  17. special_tokens_map.json +36 -0
  18. tokenizer.json +0 -0
  19. tokenizer_config.json +413 -0
chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{% for message in messages %}{{'<|' + message['role'] + '|>' + '\n' + message['content'] + '<|end|>\n' }}{% endfor %}{% if add_generation_prompt and messages[-1]['role'] != 'assistant' %}{{- '<|assistant|>\n' -}}{% endif %}"
3
+ }
config.json ADDED
The diff for this file is too large to render. See raw diff
 
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 32000,
6
+ "transformers_version": "4.45.0"
7
+ }
openvino_detokenizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a60c90df70041d2a3db95701b3bc410f557a9d2568b2055ce20b3003c778c3f9
3
+ size 340120
openvino_detokenizer.xml ADDED
@@ -0,0 +1,416 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="detokenizer" version="11">
3
+ <layers>
4
+ <layer id="0" name="Parameter_122" type="Parameter" version="opset1">
5
+ <data shape="?,?" element_type="i64" />
6
+ <output>
7
+ <port id="0" precision="I64" names="Parameter_122">
8
+ <dim>-1</dim>
9
+ <dim>-1</dim>
10
+ </port>
11
+ </output>
12
+ </layer>
13
+ <layer id="1" name="Convert_149" type="Convert" version="opset1">
14
+ <data destination_type="i32" />
15
+ <input>
16
+ <port id="0" precision="I64">
17
+ <dim>-1</dim>
18
+ <dim>-1</dim>
19
+ </port>
20
+ </input>
21
+ <output>
22
+ <port id="1" precision="I32">
23
+ <dim>-1</dim>
24
+ <dim>-1</dim>
25
+ </port>
26
+ </output>
27
+ </layer>
28
+ <layer id="2" name="Constant_89" type="Const" version="opset1">
29
+ <data element_type="u8" shape="339905" offset="0" size="339905" />
30
+ <output>
31
+ <port id="0" precision="U8">
32
+ <dim>339905</dim>
33
+ </port>
34
+ </output>
35
+ </layer>
36
+ <layer id="3" name="StringTensorUnpack_90" type="StringTensorUnpack" version="extension">
37
+ <data mode="begins_ends" />
38
+ <input>
39
+ <port id="0" precision="U8">
40
+ <dim>339905</dim>
41
+ </port>
42
+ </input>
43
+ <output>
44
+ <port id="1" precision="I32">
45
+ <dim>-1</dim>
46
+ </port>
47
+ <port id="2" precision="I32">
48
+ <dim>-1</dim>
49
+ </port>
50
+ <port id="3" precision="U8">
51
+ <dim>-1</dim>
52
+ </port>
53
+ </output>
54
+ </layer>
55
+ <layer id="4" name="Constant_126" type="Const" version="opset1">
56
+ <data element_type="i32" shape="47" offset="339905" size="188" />
57
+ <output>
58
+ <port id="0" precision="I32">
59
+ <dim>47</dim>
60
+ </port>
61
+ </output>
62
+ </layer>
63
+ <layer id="5" name="Constant_124" type="Const" version="opset1">
64
+ <data element_type="i32" shape="1" offset="340093" size="4" />
65
+ <output>
66
+ <port id="0" precision="I32">
67
+ <dim>1</dim>
68
+ </port>
69
+ </output>
70
+ </layer>
71
+ <layer id="6" name="Constant_123" type="Const" version="opset1">
72
+ <data element_type="i32" shape="1" offset="340097" size="4" />
73
+ <output>
74
+ <port id="0" precision="I32">
75
+ <dim>1</dim>
76
+ </port>
77
+ </output>
78
+ </layer>
79
+ <layer id="7" name="Constant_125" type="Const" version="opset1">
80
+ <data element_type="i32" shape="1" offset="340101" size="4" />
81
+ <output>
82
+ <port id="0" precision="I32">
83
+ <dim>1</dim>
84
+ </port>
85
+ </output>
86
+ </layer>
87
+ <layer id="8" name="Constant_128" type="Const" version="opset1">
88
+ <data element_type="i64" shape="1" offset="340105" size="8" />
89
+ <output>
90
+ <port id="0" precision="I64">
91
+ <dim>1</dim>
92
+ </port>
93
+ </output>
94
+ </layer>
95
+ <layer id="9" name="Slice_127" type="Slice" version="opset8">
96
+ <input>
97
+ <port id="0" precision="I32">
98
+ <dim>47</dim>
99
+ </port>
100
+ <port id="1" precision="I32">
101
+ <dim>1</dim>
102
+ </port>
103
+ <port id="2" precision="I32">
104
+ <dim>1</dim>
105
+ </port>
106
+ <port id="3" precision="I32">
107
+ <dim>1</dim>
108
+ </port>
109
+ <port id="4" precision="I64">
110
+ <dim>1</dim>
111
+ </port>
112
+ </input>
113
+ <output>
114
+ <port id="5" precision="I32">
115
+ <dim>47</dim>
116
+ </port>
117
+ </output>
118
+ </layer>
119
+ <layer id="10" name="VocabDecoder_129" type="VocabDecoder" version="extension">
120
+ <data skip_tokens="" />
121
+ <input>
122
+ <port id="0" precision="I32">
123
+ <dim>-1</dim>
124
+ <dim>-1</dim>
125
+ </port>
126
+ <port id="1" precision="I32">
127
+ <dim>-1</dim>
128
+ </port>
129
+ <port id="2" precision="I32">
130
+ <dim>-1</dim>
131
+ </port>
132
+ <port id="3" precision="U8">
133
+ <dim>-1</dim>
134
+ </port>
135
+ <port id="4" precision="I32">
136
+ <dim>47</dim>
137
+ </port>
138
+ </input>
139
+ <output>
140
+ <port id="5" precision="I32">
141
+ <dim>-1</dim>
142
+ </port>
143
+ <port id="6" precision="I32">
144
+ <dim>-1</dim>
145
+ </port>
146
+ <port id="7" precision="I32">
147
+ <dim>-1</dim>
148
+ </port>
149
+ <port id="8" precision="I32">
150
+ <dim>-1</dim>
151
+ </port>
152
+ <port id="9" precision="U8">
153
+ <dim>-1</dim>
154
+ </port>
155
+ </output>
156
+ </layer>
157
+ <layer id="11" name="Constant_131" type="Const" version="opset1">
158
+ <data element_type="u8" shape="3" offset="340113" size="3" />
159
+ <output>
160
+ <port id="0" precision="U8">
161
+ <dim>3</dim>
162
+ </port>
163
+ </output>
164
+ </layer>
165
+ <layer id="12" name="Constant_133" type="Const" version="opset1">
166
+ <data element_type="u8" shape="1" offset="340116" size="1" />
167
+ <output>
168
+ <port id="0" precision="U8">
169
+ <dim>1</dim>
170
+ </port>
171
+ </output>
172
+ </layer>
173
+ <layer id="13" name="RegexNormalization_134" type="RegexNormalization" version="extension">
174
+ <data global_replace="true" />
175
+ <input>
176
+ <port id="0" precision="I32">
177
+ <dim>-1</dim>
178
+ </port>
179
+ <port id="1" precision="I32">
180
+ <dim>-1</dim>
181
+ </port>
182
+ <port id="2" precision="U8">
183
+ <dim>-1</dim>
184
+ </port>
185
+ <port id="3" precision="U8">
186
+ <dim>3</dim>
187
+ </port>
188
+ <port id="4" precision="U8">
189
+ <dim>1</dim>
190
+ </port>
191
+ </input>
192
+ <output>
193
+ <port id="5" precision="I32">
194
+ <dim>-1</dim>
195
+ </port>
196
+ <port id="6" precision="I32">
197
+ <dim>-1</dim>
198
+ </port>
199
+ <port id="7" precision="U8">
200
+ <dim>-1</dim>
201
+ </port>
202
+ </output>
203
+ </layer>
204
+ <layer id="14" name="ByteFallback_135" type="ByteFallback" version="extension">
205
+ <input>
206
+ <port id="0" precision="I32">
207
+ <dim>-1</dim>
208
+ </port>
209
+ <port id="1" precision="I32">
210
+ <dim>-1</dim>
211
+ </port>
212
+ <port id="2" precision="U8">
213
+ <dim>-1</dim>
214
+ </port>
215
+ </input>
216
+ <output>
217
+ <port id="3" precision="I32">
218
+ <dim>-1</dim>
219
+ </port>
220
+ <port id="4" precision="I32">
221
+ <dim>-1</dim>
222
+ </port>
223
+ <port id="5" precision="U8">
224
+ <dim>-1</dim>
225
+ </port>
226
+ </output>
227
+ </layer>
228
+ <layer id="15" name="FuzeRagged_136" type="FuzeRagged" version="extension">
229
+ <input>
230
+ <port id="0" precision="I32">
231
+ <dim>-1</dim>
232
+ </port>
233
+ <port id="1" precision="I32">
234
+ <dim>-1</dim>
235
+ </port>
236
+ <port id="2" precision="I32">
237
+ <dim>-1</dim>
238
+ </port>
239
+ <port id="3" precision="I32">
240
+ <dim>-1</dim>
241
+ </port>
242
+ </input>
243
+ <output>
244
+ <port id="4" precision="I32">
245
+ <dim>-1</dim>
246
+ </port>
247
+ <port id="5" precision="I32">
248
+ <dim>-1</dim>
249
+ </port>
250
+ </output>
251
+ </layer>
252
+ <layer id="16" name="Constant_138" type="Const" version="opset1">
253
+ <data element_type="u8" shape="2" offset="340117" size="2" />
254
+ <output>
255
+ <port id="0" precision="U8">
256
+ <dim>2</dim>
257
+ </port>
258
+ </output>
259
+ </layer>
260
+ <layer id="17" name="Constant_140" type="Const" version="opset1">
261
+ <data element_type="u8" shape="0" offset="340119" size="1" />
262
+ <output>
263
+ <port id="0" precision="U8">
264
+ <dim>0</dim>
265
+ </port>
266
+ </output>
267
+ </layer>
268
+ <layer id="18" name="RegexNormalization_141" type="RegexNormalization" version="extension">
269
+ <data global_replace="true" />
270
+ <input>
271
+ <port id="0" precision="I32">
272
+ <dim>-1</dim>
273
+ </port>
274
+ <port id="1" precision="I32">
275
+ <dim>-1</dim>
276
+ </port>
277
+ <port id="2" precision="U8">
278
+ <dim>-1</dim>
279
+ </port>
280
+ <port id="3" precision="U8">
281
+ <dim>2</dim>
282
+ </port>
283
+ <port id="4" precision="U8">
284
+ <dim>0</dim>
285
+ </port>
286
+ </input>
287
+ <output>
288
+ <port id="5" precision="I32">
289
+ <dim>-1</dim>
290
+ </port>
291
+ <port id="6" precision="I32">
292
+ <dim>-1</dim>
293
+ </port>
294
+ <port id="7" precision="U8">
295
+ <dim>-1</dim>
296
+ </port>
297
+ </output>
298
+ </layer>
299
+ <layer id="19" name="UTF8Validate_142" type="UTF8Validate" version="extension">
300
+ <data replace_mode="true" />
301
+ <input>
302
+ <port id="0" precision="I32">
303
+ <dim>-1</dim>
304
+ </port>
305
+ <port id="1" precision="I32">
306
+ <dim>-1</dim>
307
+ </port>
308
+ <port id="2" precision="U8">
309
+ <dim>-1</dim>
310
+ </port>
311
+ </input>
312
+ <output>
313
+ <port id="3" precision="I32">
314
+ <dim>-1</dim>
315
+ </port>
316
+ <port id="4" precision="I32">
317
+ <dim>-1</dim>
318
+ </port>
319
+ <port id="5" precision="U8">
320
+ <dim>-1</dim>
321
+ </port>
322
+ </output>
323
+ </layer>
324
+ <layer id="20" name="StringTensorPack_143" type="StringTensorPack" version="extension">
325
+ <data mode="begins_ends" />
326
+ <input>
327
+ <port id="0" precision="I32">
328
+ <dim>-1</dim>
329
+ </port>
330
+ <port id="1" precision="I32">
331
+ <dim>-1</dim>
332
+ </port>
333
+ <port id="2" precision="U8">
334
+ <dim>-1</dim>
335
+ </port>
336
+ </input>
337
+ <output>
338
+ <port id="3" precision="STRING" names="string_output">
339
+ <dim>-1</dim>
340
+ </port>
341
+ </output>
342
+ </layer>
343
+ <layer id="21" name="Result_144" type="Result" version="opset1">
344
+ <input>
345
+ <port id="0" precision="STRING">
346
+ <dim>-1</dim>
347
+ </port>
348
+ </input>
349
+ </layer>
350
+ </layers>
351
+ <edges>
352
+ <edge from-layer="0" from-port="0" to-layer="1" to-port="0" />
353
+ <edge from-layer="1" from-port="1" to-layer="10" to-port="0" />
354
+ <edge from-layer="2" from-port="0" to-layer="3" to-port="0" />
355
+ <edge from-layer="3" from-port="3" to-layer="10" to-port="3" />
356
+ <edge from-layer="3" from-port="2" to-layer="10" to-port="2" />
357
+ <edge from-layer="3" from-port="1" to-layer="10" to-port="1" />
358
+ <edge from-layer="4" from-port="0" to-layer="9" to-port="0" />
359
+ <edge from-layer="5" from-port="0" to-layer="9" to-port="1" />
360
+ <edge from-layer="6" from-port="0" to-layer="9" to-port="2" />
361
+ <edge from-layer="7" from-port="0" to-layer="9" to-port="3" />
362
+ <edge from-layer="8" from-port="0" to-layer="9" to-port="4" />
363
+ <edge from-layer="9" from-port="5" to-layer="10" to-port="4" />
364
+ <edge from-layer="10" from-port="7" to-layer="13" to-port="0" />
365
+ <edge from-layer="10" from-port="8" to-layer="13" to-port="1" />
366
+ <edge from-layer="10" from-port="9" to-layer="13" to-port="2" />
367
+ <edge from-layer="10" from-port="6" to-layer="15" to-port="1" />
368
+ <edge from-layer="10" from-port="5" to-layer="15" to-port="0" />
369
+ <edge from-layer="11" from-port="0" to-layer="13" to-port="3" />
370
+ <edge from-layer="12" from-port="0" to-layer="13" to-port="4" />
371
+ <edge from-layer="13" from-port="6" to-layer="14" to-port="1" />
372
+ <edge from-layer="13" from-port="7" to-layer="14" to-port="2" />
373
+ <edge from-layer="13" from-port="5" to-layer="14" to-port="0" />
374
+ <edge from-layer="14" from-port="3" to-layer="15" to-port="2" />
375
+ <edge from-layer="14" from-port="4" to-layer="15" to-port="3" />
376
+ <edge from-layer="14" from-port="5" to-layer="18" to-port="2" />
377
+ <edge from-layer="15" from-port="4" to-layer="18" to-port="0" />
378
+ <edge from-layer="15" from-port="5" to-layer="18" to-port="1" />
379
+ <edge from-layer="16" from-port="0" to-layer="18" to-port="3" />
380
+ <edge from-layer="17" from-port="0" to-layer="18" to-port="4" />
381
+ <edge from-layer="18" from-port="5" to-layer="19" to-port="0" />
382
+ <edge from-layer="18" from-port="6" to-layer="19" to-port="1" />
383
+ <edge from-layer="18" from-port="7" to-layer="19" to-port="2" />
384
+ <edge from-layer="19" from-port="3" to-layer="20" to-port="0" />
385
+ <edge from-layer="19" from-port="4" to-layer="20" to-port="1" />
386
+ <edge from-layer="19" from-port="5" to-layer="20" to-port="2" />
387
+ <edge from-layer="20" from-port="3" to-layer="21" to-port="0" />
388
+ </edges>
389
+ <rt_info>
390
+ <add_attention_mask value="True" />
391
+ <add_prefix_space />
392
+ <add_special_tokens value="True" />
393
+ <bos_token_id value="1" />
394
+ <chat_template value="{% for message in messages %}{{'&lt;|' + message['role'] + '|>' + '&#10;' + message['content'] + '&lt;|end|>&#10;' }}{% endfor %}{% if add_generation_prompt and messages[-1]['role'] != 'assistant' %}{{- '&lt;|assistant|>&#10;' -}}{% endif %}" />
395
+ <clean_up_tokenization_spaces />
396
+ <detokenizer_input_type value="i64" />
397
+ <eos_token_id value="32000" />
398
+ <handle_special_tokens_with_re value="False" />
399
+ <number_of_inputs value="1" />
400
+ <openvino_tokenizers_version value="2025.0.0.0rc2" />
401
+ <openvino_version value="2025.0.0rc2" />
402
+ <original_tokenizer_class value="&lt;class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>" />
403
+ <pad_token_id value="32000" />
404
+ <sentencepiece_version value="0.2.0" />
405
+ <skip_special_tokens value="True" />
406
+ <streaming_detokenizer value="False" />
407
+ <tiktoken_version value="0.8.0" />
408
+ <tokenizer_output_type value="i64" />
409
+ <tokenizers_version value="0.20.1" />
410
+ <transformers_version value="4.45.0" />
411
+ <use_max_padding value="False" />
412
+ <use_sentencepiece_backend value="True" />
413
+ <utf8_replace_mode value="replace" />
414
+ <with_detokenizer value="True" />
415
+ </rt_info>
416
+ </net>
openvino_language_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9305e4cb77c5c7576bf5b7458e07fe3aaa2012725b75d0b48a04efc4864e7774
3
+ size 7445158228
openvino_language_model.xml ADDED
The diff for this file is too large to render. See raw diff
 
openvino_text_embeddings_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc0fc37f3cd84cf9313a7525875c2a02485ea1aa662d8cb3f4c807ddffd3357f
3
+ size 197001220
openvino_text_embeddings_model.xml ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="Model9" version="11">
3
+ <layers>
4
+ <layer id="0" name="input" type="Parameter" version="opset1">
5
+ <data shape="?,?" element_type="i64" />
6
+ <output>
7
+ <port id="0" precision="I64" names="input">
8
+ <dim>-1</dim>
9
+ <dim>-1</dim>
10
+ </port>
11
+ </output>
12
+ </layer>
13
+ <layer id="1" name="self.weight_compressed" type="Const" version="opset1">
14
+ <data element_type="f16" shape="32064, 3072" offset="0" size="197001216" />
15
+ <output>
16
+ <port id="0" precision="FP16" names="self.weight">
17
+ <dim>32064</dim>
18
+ <dim>3072</dim>
19
+ </port>
20
+ </output>
21
+ </layer>
22
+ <layer id="2" name="self.weight" type="Convert" version="opset1">
23
+ <data destination_type="f32" />
24
+ <rt_info>
25
+ <attribute name="decompression" version="0" />
26
+ </rt_info>
27
+ <input>
28
+ <port id="0" precision="FP16">
29
+ <dim>32064</dim>
30
+ <dim>3072</dim>
31
+ </port>
32
+ </input>
33
+ <output>
34
+ <port id="1" precision="FP32">
35
+ <dim>32064</dim>
36
+ <dim>3072</dim>
37
+ </port>
38
+ </output>
39
+ </layer>
40
+ <layer id="3" name="aten::embedding/Convert" type="Convert" version="opset1">
41
+ <data destination_type="i32" />
42
+ <input>
43
+ <port id="0" precision="I64">
44
+ <dim>-1</dim>
45
+ <dim>-1</dim>
46
+ </port>
47
+ </input>
48
+ <output>
49
+ <port id="1" precision="I32">
50
+ <dim>-1</dim>
51
+ <dim>-1</dim>
52
+ </port>
53
+ </output>
54
+ </layer>
55
+ <layer id="4" name="aten::embedding/Constant" type="Const" version="opset1">
56
+ <data element_type="i32" shape="" offset="197001216" size="4" />
57
+ <output>
58
+ <port id="0" precision="I32" />
59
+ </output>
60
+ </layer>
61
+ <layer id="5" name="aten::embedding/Gather" type="Gather" version="opset8">
62
+ <data batch_dims="0" />
63
+ <input>
64
+ <port id="0" precision="FP32">
65
+ <dim>32064</dim>
66
+ <dim>3072</dim>
67
+ </port>
68
+ <port id="1" precision="I32">
69
+ <dim>-1</dim>
70
+ <dim>-1</dim>
71
+ </port>
72
+ <port id="2" precision="I32" />
73
+ </input>
74
+ <output>
75
+ <port id="3" precision="FP32" names="inputs_embeds">
76
+ <dim>-1</dim>
77
+ <dim>-1</dim>
78
+ <dim>3072</dim>
79
+ </port>
80
+ </output>
81
+ </layer>
82
+ <layer id="6" name="Result_132780" type="Result" version="opset1">
83
+ <input>
84
+ <port id="0" precision="FP32">
85
+ <dim>-1</dim>
86
+ <dim>-1</dim>
87
+ <dim>3072</dim>
88
+ </port>
89
+ </input>
90
+ </layer>
91
+ </layers>
92
+ <edges>
93
+ <edge from-layer="0" from-port="0" to-layer="3" to-port="0" />
94
+ <edge from-layer="1" from-port="0" to-layer="2" to-port="0" />
95
+ <edge from-layer="2" from-port="1" to-layer="5" to-port="0" />
96
+ <edge from-layer="3" from-port="1" to-layer="5" to-port="1" />
97
+ <edge from-layer="4" from-port="0" to-layer="5" to-port="2" />
98
+ <edge from-layer="5" from-port="3" to-layer="6" to-port="0" />
99
+ </edges>
100
+ <rt_info>
101
+ <Runtime_version value="2025.0.0-17933-815af98acd8-releases/2025/0" />
102
+ <conversion_parameters>
103
+ <framework value="pytorch" />
104
+ <is_python_object value="True" />
105
+ </conversion_parameters>
106
+ <optimum>
107
+ <optimum_intel_version value="1.22.0.dev0+753f84d" />
108
+ <optimum_version value="1.24.0.dev0" />
109
+ <pytorch_version value="2.5.0+cpu" />
110
+ <transformers_version value="4.45.0" />
111
+ </optimum>
112
+ </rt_info>
113
+ </net>
openvino_tokenizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c1510cca0328e390459199fa46fde85bc7a72c42a3156d44a9c5b3975daca20
3
+ size 1300299
openvino_tokenizer.xml ADDED
@@ -0,0 +1,835 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="tokenizer" version="11">
3
+ <layers>
4
+ <layer id="0" name="Parameter_1" type="Parameter" version="opset1">
5
+ <data shape="?" element_type="string" />
6
+ <output>
7
+ <port id="0" precision="STRING" names="Parameter_1">
8
+ <dim>-1</dim>
9
+ </port>
10
+ </output>
11
+ </layer>
12
+ <layer id="1" name="Constant_106" type="Const" version="opset1">
13
+ <data element_type="i32" shape="" offset="0" size="4" />
14
+ <output>
15
+ <port id="0" precision="I32" />
16
+ </output>
17
+ </layer>
18
+ <layer id="2" name="Constant_107" type="Const" version="opset1">
19
+ <data element_type="i32" shape="" offset="4" size="4" />
20
+ <output>
21
+ <port id="0" precision="I32" />
22
+ </output>
23
+ </layer>
24
+ <layer id="3" name="Constant_108" type="Const" version="opset1">
25
+ <data element_type="i32" shape="1" offset="4" size="4" />
26
+ <output>
27
+ <port id="0" precision="I32">
28
+ <dim>1</dim>
29
+ </port>
30
+ </output>
31
+ </layer>
32
+ <layer id="4" name="Constant_7" type="Const" version="opset1">
33
+ <data element_type="i64" shape="" offset="8" size="8" />
34
+ <output>
35
+ <port id="0" precision="I64" />
36
+ </output>
37
+ </layer>
38
+ <layer id="5" name="StringTensorUnpack_2" type="StringTensorUnpack" version="extension">
39
+ <data mode="begins_ends" />
40
+ <input>
41
+ <port id="0" precision="STRING">
42
+ <dim>-1</dim>
43
+ </port>
44
+ </input>
45
+ <output>
46
+ <port id="1" precision="I32">
47
+ <dim>-1</dim>
48
+ </port>
49
+ <port id="2" precision="I32">
50
+ <dim>-1</dim>
51
+ </port>
52
+ <port id="3" precision="U8">
53
+ <dim>-1</dim>
54
+ </port>
55
+ </output>
56
+ </layer>
57
+ <layer id="6" name="ShapeOf_3" type="ShapeOf" version="opset3">
58
+ <data output_type="i64" />
59
+ <input>
60
+ <port id="0" precision="I32">
61
+ <dim>-1</dim>
62
+ </port>
63
+ </input>
64
+ <output>
65
+ <port id="1" precision="I64">
66
+ <dim>1</dim>
67
+ </port>
68
+ </output>
69
+ </layer>
70
+ <layer id="7" name="Constant_4" type="Const" version="opset1">
71
+ <data element_type="i64" shape="" offset="8" size="8" />
72
+ <output>
73
+ <port id="0" precision="I64" />
74
+ </output>
75
+ </layer>
76
+ <layer id="8" name="Constant_5" type="Const" version="opset1">
77
+ <data element_type="i64" shape="" offset="8" size="8" />
78
+ <output>
79
+ <port id="0" precision="I64" />
80
+ </output>
81
+ </layer>
82
+ <layer id="9" name="Gather_6" type="Gather" version="opset8">
83
+ <data batch_dims="0" />
84
+ <input>
85
+ <port id="0" precision="I64">
86
+ <dim>1</dim>
87
+ </port>
88
+ <port id="1" precision="I64" />
89
+ <port id="2" precision="I64" />
90
+ </input>
91
+ <output>
92
+ <port id="3" precision="I64" />
93
+ </output>
94
+ </layer>
95
+ <layer id="10" name="Constant_8" type="Const" version="opset1">
96
+ <data element_type="i64" shape="" offset="16" size="8" />
97
+ <output>
98
+ <port id="0" precision="I64" />
99
+ </output>
100
+ </layer>
101
+ <layer id="11" name="Range_9" type="Range" version="opset4">
102
+ <data output_type="i32" />
103
+ <input>
104
+ <port id="0" precision="I64" />
105
+ <port id="1" precision="I64" />
106
+ <port id="2" precision="I64" />
107
+ </input>
108
+ <output>
109
+ <port id="3" precision="I32">
110
+ <dim>-1</dim>
111
+ </port>
112
+ </output>
113
+ </layer>
114
+ <layer id="12" name="Constant_10" type="Const" version="opset1">
115
+ <data element_type="i64" shape="" offset="16" size="8" />
116
+ <output>
117
+ <port id="0" precision="I64" />
118
+ </output>
119
+ </layer>
120
+ <layer id="13" name="Constant_11" type="Const" version="opset1">
121
+ <data element_type="i64" shape="" offset="16" size="8" />
122
+ <output>
123
+ <port id="0" precision="I64" />
124
+ </output>
125
+ </layer>
126
+ <layer id="14" name="Add_12" type="Add" version="opset1">
127
+ <data auto_broadcast="numpy" />
128
+ <input>
129
+ <port id="0" precision="I64" />
130
+ <port id="1" precision="I64" />
131
+ </input>
132
+ <output>
133
+ <port id="2" precision="I64" />
134
+ </output>
135
+ </layer>
136
+ <layer id="15" name="Constant_13" type="Const" version="opset1">
137
+ <data element_type="i64" shape="" offset="16" size="8" />
138
+ <output>
139
+ <port id="0" precision="I64" />
140
+ </output>
141
+ </layer>
142
+ <layer id="16" name="Range_14" type="Range" version="opset4">
143
+ <data output_type="i32" />
144
+ <input>
145
+ <port id="0" precision="I64" />
146
+ <port id="1" precision="I64" />
147
+ <port id="2" precision="I64" />
148
+ </input>
149
+ <output>
150
+ <port id="3" precision="I32">
151
+ <dim>-1</dim>
152
+ </port>
153
+ </output>
154
+ </layer>
155
+ <layer id="17" name="Constant_76" type="Const" version="opset1">
156
+ <data element_type="u8" shape="1350" offset="24" size="1350" />
157
+ <output>
158
+ <port id="0" precision="U8">
159
+ <dim>1350</dim>
160
+ </port>
161
+ </output>
162
+ </layer>
163
+ <layer id="18" name="SpecialTokensSplit_77" type="SpecialTokensSplit" version="extension">
164
+ <input>
165
+ <port id="0" precision="I32">
166
+ <dim>-1</dim>
167
+ </port>
168
+ <port id="1" precision="I32">
169
+ <dim>-1</dim>
170
+ </port>
171
+ <port id="2" precision="I32">
172
+ <dim>-1</dim>
173
+ </port>
174
+ <port id="3" precision="I32">
175
+ <dim>-1</dim>
176
+ </port>
177
+ <port id="4" precision="U8">
178
+ <dim>-1</dim>
179
+ </port>
180
+ <port id="5" precision="U8">
181
+ <dim>1350</dim>
182
+ </port>
183
+ </input>
184
+ <output>
185
+ <port id="6" precision="I32">
186
+ <dim>-1</dim>
187
+ </port>
188
+ <port id="7" precision="I32">
189
+ <dim>-1</dim>
190
+ </port>
191
+ <port id="8" precision="I32">
192
+ <dim>-1</dim>
193
+ </port>
194
+ <port id="9" precision="I32">
195
+ <dim>-1</dim>
196
+ </port>
197
+ <port id="10" precision="U8">
198
+ <dim>-1</dim>
199
+ </port>
200
+ <port id="11" precision="BOOL">
201
+ <dim>-1</dim>
202
+ </port>
203
+ </output>
204
+ </layer>
205
+ <layer id="19" name="Constant_79" type="Const" version="opset1">
206
+ <data element_type="u8" shape="13" offset="1374" size="13" />
207
+ <output>
208
+ <port id="0" precision="U8">
209
+ <dim>13</dim>
210
+ </port>
211
+ </output>
212
+ </layer>
213
+ <layer id="20" name="Constant_81" type="Const" version="opset1">
214
+ <data element_type="u8" shape="5" offset="1387" size="5" />
215
+ <output>
216
+ <port id="0" precision="U8">
217
+ <dim>5</dim>
218
+ </port>
219
+ </output>
220
+ </layer>
221
+ <layer id="21" name="RegexNormalization_82" type="RegexNormalization" version="extension">
222
+ <data global_replace="true" />
223
+ <input>
224
+ <port id="0" precision="I32">
225
+ <dim>-1</dim>
226
+ </port>
227
+ <port id="1" precision="I32">
228
+ <dim>-1</dim>
229
+ </port>
230
+ <port id="2" precision="U8">
231
+ <dim>-1</dim>
232
+ </port>
233
+ <port id="3" precision="BOOL">
234
+ <dim>-1</dim>
235
+ </port>
236
+ <port id="4" precision="U8">
237
+ <dim>13</dim>
238
+ </port>
239
+ <port id="5" precision="U8">
240
+ <dim>5</dim>
241
+ </port>
242
+ </input>
243
+ <output>
244
+ <port id="6" precision="I32">
245
+ <dim>-1</dim>
246
+ </port>
247
+ <port id="7" precision="I32">
248
+ <dim>-1</dim>
249
+ </port>
250
+ <port id="8" precision="U8">
251
+ <dim>-1</dim>
252
+ </port>
253
+ <port id="9" precision="BOOL">
254
+ <dim>-1</dim>
255
+ </port>
256
+ </output>
257
+ </layer>
258
+ <layer id="22" name="Constant_84" type="Const" version="opset1">
259
+ <data element_type="u8" shape="1" offset="1392" size="1" />
260
+ <output>
261
+ <port id="0" precision="U8">
262
+ <dim>1</dim>
263
+ </port>
264
+ </output>
265
+ </layer>
266
+ <layer id="23" name="Constant_86" type="Const" version="opset1">
267
+ <data element_type="u8" shape="3" offset="1393" size="3" />
268
+ <output>
269
+ <port id="0" precision="U8">
270
+ <dim>3</dim>
271
+ </port>
272
+ </output>
273
+ </layer>
274
+ <layer id="24" name="RegexNormalization_87" type="RegexNormalization" version="extension">
275
+ <data global_replace="true" />
276
+ <input>
277
+ <port id="0" precision="I32">
278
+ <dim>-1</dim>
279
+ </port>
280
+ <port id="1" precision="I32">
281
+ <dim>-1</dim>
282
+ </port>
283
+ <port id="2" precision="U8">
284
+ <dim>-1</dim>
285
+ </port>
286
+ <port id="3" precision="BOOL">
287
+ <dim>-1</dim>
288
+ </port>
289
+ <port id="4" precision="U8">
290
+ <dim>1</dim>
291
+ </port>
292
+ <port id="5" precision="U8">
293
+ <dim>3</dim>
294
+ </port>
295
+ </input>
296
+ <output>
297
+ <port id="6" precision="I32">
298
+ <dim>-1</dim>
299
+ </port>
300
+ <port id="7" precision="I32">
301
+ <dim>-1</dim>
302
+ </port>
303
+ <port id="8" precision="U8">
304
+ <dim>-1</dim>
305
+ </port>
306
+ <port id="9" precision="BOOL">
307
+ <dim>-1</dim>
308
+ </port>
309
+ </output>
310
+ </layer>
311
+ <layer id="25" name="Constant_89" type="Const" version="opset1">
312
+ <data element_type="u8" shape="339905" offset="1396" size="339905" />
313
+ <output>
314
+ <port id="0" precision="U8">
315
+ <dim>339905</dim>
316
+ </port>
317
+ </output>
318
+ </layer>
319
+ <layer id="26" name="StringTensorUnpack_90" type="StringTensorUnpack" version="extension">
320
+ <data mode="begins_ends" />
321
+ <input>
322
+ <port id="0" precision="U8">
323
+ <dim>339905</dim>
324
+ </port>
325
+ </input>
326
+ <output>
327
+ <port id="1" precision="I32">
328
+ <dim>-1</dim>
329
+ </port>
330
+ <port id="2" precision="I32">
331
+ <dim>-1</dim>
332
+ </port>
333
+ <port id="3" precision="U8">
334
+ <dim>-1</dim>
335
+ </port>
336
+ </output>
337
+ </layer>
338
+ <layer id="27" name="Constant_95" type="Const" version="opset1">
339
+ <data element_type="u8" shape="499127" offset="341301" size="499127" />
340
+ <output>
341
+ <port id="0" precision="U8">
342
+ <dim>499127</dim>
343
+ </port>
344
+ </output>
345
+ </layer>
346
+ <layer id="28" name="StringTensorUnpack_96" type="StringTensorUnpack" version="extension">
347
+ <data mode="begins_ends" />
348
+ <input>
349
+ <port id="0" precision="U8">
350
+ <dim>499127</dim>
351
+ </port>
352
+ </input>
353
+ <output>
354
+ <port id="1" precision="I32">
355
+ <dim>-1</dim>
356
+ </port>
357
+ <port id="2" precision="I32">
358
+ <dim>-1</dim>
359
+ </port>
360
+ <port id="3" precision="U8">
361
+ <dim>-1</dim>
362
+ </port>
363
+ </output>
364
+ </layer>
365
+ <layer id="29" name="Constant_98" type="Const" version="opset1">
366
+ <data element_type="u8" shape="412810" offset="840428" size="412810" />
367
+ <output>
368
+ <port id="0" precision="U8">
369
+ <dim>412810</dim>
370
+ </port>
371
+ </output>
372
+ </layer>
373
+ <layer id="30" name="StringTensorUnpack_99" type="StringTensorUnpack" version="extension">
374
+ <data mode="begins_ends" />
375
+ <input>
376
+ <port id="0" precision="U8">
377
+ <dim>412810</dim>
378
+ </port>
379
+ </input>
380
+ <output>
381
+ <port id="1" precision="I32">
382
+ <dim>-1</dim>
383
+ </port>
384
+ <port id="2" precision="I32">
385
+ <dim>-1</dim>
386
+ </port>
387
+ <port id="3" precision="U8">
388
+ <dim>-1</dim>
389
+ </port>
390
+ </output>
391
+ </layer>
392
+ <layer id="31" name="Constant_92" type="Const" version="opset1">
393
+ <data element_type="u8" shape="39809" offset="1253238" size="39809" />
394
+ <output>
395
+ <port id="0" precision="U8">
396
+ <dim>39809</dim>
397
+ </port>
398
+ </output>
399
+ </layer>
400
+ <layer id="32" name="StringTensorUnpack_93" type="StringTensorUnpack" version="extension">
401
+ <data mode="begins_ends" />
402
+ <input>
403
+ <port id="0" precision="U8">
404
+ <dim>39809</dim>
405
+ </port>
406
+ </input>
407
+ <output>
408
+ <port id="1" precision="I32">
409
+ <dim>-1</dim>
410
+ </port>
411
+ <port id="2" precision="I32">
412
+ <dim>-1</dim>
413
+ </port>
414
+ <port id="3" precision="U8">
415
+ <dim>-1</dim>
416
+ </port>
417
+ </output>
418
+ </layer>
419
+ <layer id="33" name="Constant_100" type="Const" version="opset1">
420
+ <data element_type="i32" shape="1811" offset="1293047" size="7244" />
421
+ <output>
422
+ <port id="0" precision="I32">
423
+ <dim>1811</dim>
424
+ </port>
425
+ </output>
426
+ </layer>
427
+ <layer id="34" name="BPETokenizer_101" type="BPETokenizer" version="extension">
428
+ <data unk_token="&lt;unk>" fuse_unk="true" suffix_indicator="" end_suffix="" byte_fallback="true" cache_capacity="20000" />
429
+ <input>
430
+ <port id="0" precision="I32">
431
+ <dim>-1</dim>
432
+ </port>
433
+ <port id="1" precision="I32">
434
+ <dim>-1</dim>
435
+ </port>
436
+ <port id="2" precision="I32">
437
+ <dim>-1</dim>
438
+ </port>
439
+ <port id="3" precision="I32">
440
+ <dim>-1</dim>
441
+ </port>
442
+ <port id="4" precision="U8">
443
+ <dim>-1</dim>
444
+ </port>
445
+ <port id="5" precision="I32">
446
+ <dim>-1</dim>
447
+ </port>
448
+ <port id="6" precision="I32">
449
+ <dim>-1</dim>
450
+ </port>
451
+ <port id="7" precision="U8">
452
+ <dim>-1</dim>
453
+ </port>
454
+ <port id="8" precision="I32">
455
+ <dim>-1</dim>
456
+ </port>
457
+ <port id="9" precision="I32">
458
+ <dim>-1</dim>
459
+ </port>
460
+ <port id="10" precision="U8">
461
+ <dim>-1</dim>
462
+ </port>
463
+ <port id="11" precision="I32">
464
+ <dim>-1</dim>
465
+ </port>
466
+ <port id="12" precision="I32">
467
+ <dim>-1</dim>
468
+ </port>
469
+ <port id="13" precision="U8">
470
+ <dim>-1</dim>
471
+ </port>
472
+ <port id="14" precision="I32">
473
+ <dim>-1</dim>
474
+ </port>
475
+ <port id="15" precision="I32">
476
+ <dim>-1</dim>
477
+ </port>
478
+ <port id="16" precision="U8">
479
+ <dim>-1</dim>
480
+ </port>
481
+ <port id="17" precision="I32">
482
+ <dim>1811</dim>
483
+ </port>
484
+ </input>
485
+ <output>
486
+ <port id="18" precision="I32">
487
+ <dim>-1</dim>
488
+ </port>
489
+ <port id="19" precision="I32">
490
+ <dim>-1</dim>
491
+ </port>
492
+ <port id="20" precision="I32">
493
+ <dim>-1</dim>
494
+ </port>
495
+ </output>
496
+ </layer>
497
+ <layer id="35" name="Subtract_102" type="Subtract" version="opset1">
498
+ <data auto_broadcast="numpy" />
499
+ <input>
500
+ <port id="0" precision="I32">
501
+ <dim>-1</dim>
502
+ </port>
503
+ <port id="1" precision="I32">
504
+ <dim>-1</dim>
505
+ </port>
506
+ </input>
507
+ <output>
508
+ <port id="2" precision="I32">
509
+ <dim>-1</dim>
510
+ </port>
511
+ </output>
512
+ </layer>
513
+ <layer id="36" name="Constant_103" type="Const" version="opset1">
514
+ <data element_type="i32" shape="" offset="1300291" size="4" />
515
+ <output>
516
+ <port id="0" precision="I32" />
517
+ </output>
518
+ </layer>
519
+ <layer id="37" name="Minimum_104" type="Minimum" version="opset1">
520
+ <data auto_broadcast="numpy" />
521
+ <input>
522
+ <port id="0" precision="I32">
523
+ <dim>-1</dim>
524
+ </port>
525
+ <port id="1" precision="I32" />
526
+ </input>
527
+ <output>
528
+ <port id="2" precision="I32">
529
+ <dim>-1</dim>
530
+ </port>
531
+ </output>
532
+ </layer>
533
+ <layer id="38" name="Add_105" type="Add" version="opset1">
534
+ <data auto_broadcast="numpy" />
535
+ <input>
536
+ <port id="0" precision="I32">
537
+ <dim>-1</dim>
538
+ </port>
539
+ <port id="1" precision="I32">
540
+ <dim>-1</dim>
541
+ </port>
542
+ </input>
543
+ <output>
544
+ <port id="2" precision="I32">
545
+ <dim>-1</dim>
546
+ </port>
547
+ </output>
548
+ </layer>
549
+ <layer id="39" name="Constant_109" type="Const" version="opset1">
550
+ <data element_type="i32" shape="2" offset="8" size="8" />
551
+ <output>
552
+ <port id="0" precision="I32">
553
+ <dim>2</dim>
554
+ </port>
555
+ </output>
556
+ </layer>
557
+ <layer id="40" name="CombineSegments_110" type="CombineSegments" version="extension">
558
+ <input>
559
+ <port id="0" precision="I32" />
560
+ <port id="1" precision="I32" />
561
+ <port id="2" precision="I32">
562
+ <dim>1</dim>
563
+ </port>
564
+ <port id="3" precision="I32">
565
+ <dim>-1</dim>
566
+ </port>
567
+ <port id="4" precision="I32">
568
+ <dim>-1</dim>
569
+ </port>
570
+ <port id="5" precision="I32">
571
+ <dim>-1</dim>
572
+ </port>
573
+ <port id="6" precision="I32">
574
+ <dim>2</dim>
575
+ </port>
576
+ </input>
577
+ <output>
578
+ <port id="7" precision="I32">
579
+ <dim>-1</dim>
580
+ </port>
581
+ <port id="8" precision="I32">
582
+ <dim>-1</dim>
583
+ </port>
584
+ <port id="9" precision="I32">
585
+ <dim>-1</dim>
586
+ </port>
587
+ <port id="10" precision="I32">
588
+ <dim>-1</dim>
589
+ </port>
590
+ <port id="11" precision="I32">
591
+ <dim>-1</dim>
592
+ </port>
593
+ <port id="12" precision="I32">
594
+ <dim>-1</dim>
595
+ </port>
596
+ </output>
597
+ </layer>
598
+ <layer id="41" name="Subtract_111" type="Subtract" version="opset1">
599
+ <data auto_broadcast="numpy" />
600
+ <input>
601
+ <port id="0" precision="I32">
602
+ <dim>-1</dim>
603
+ </port>
604
+ <port id="1" precision="I32">
605
+ <dim>-1</dim>
606
+ </port>
607
+ </input>
608
+ <output>
609
+ <port id="2" precision="I32">
610
+ <dim>-1</dim>
611
+ </port>
612
+ </output>
613
+ </layer>
614
+ <layer id="42" name="Constant_112" type="Const" version="opset1">
615
+ <data element_type="i32" shape="" offset="0" size="4" />
616
+ <output>
617
+ <port id="0" precision="I32" />
618
+ </output>
619
+ </layer>
620
+ <layer id="43" name="ReduceMax_113" type="ReduceMax" version="opset1">
621
+ <data keep_dims="false" />
622
+ <input>
623
+ <port id="0" precision="I32">
624
+ <dim>-1</dim>
625
+ </port>
626
+ <port id="1" precision="I32" />
627
+ </input>
628
+ <output>
629
+ <port id="2" precision="I32" />
630
+ </output>
631
+ </layer>
632
+ <layer id="44" name="Constant_114" type="Const" version="opset1">
633
+ <data element_type="i32" shape="" offset="1300295" size="4" />
634
+ <output>
635
+ <port id="0" precision="I32" />
636
+ </output>
637
+ </layer>
638
+ <layer id="45" name="RaggedToDense_115" type="RaggedToDense" version="extension">
639
+ <data pad_right="true" />
640
+ <input>
641
+ <port id="0" precision="I32">
642
+ <dim>-1</dim>
643
+ </port>
644
+ <port id="1" precision="I32">
645
+ <dim>-1</dim>
646
+ </port>
647
+ <port id="2" precision="I32">
648
+ <dim>-1</dim>
649
+ </port>
650
+ <port id="3" precision="I32" />
651
+ <port id="4" precision="I32" />
652
+ </input>
653
+ <output>
654
+ <port id="5" precision="I32">
655
+ <dim>-1</dim>
656
+ <dim>-1</dim>
657
+ </port>
658
+ <port id="6" precision="BOOL">
659
+ <dim>-1</dim>
660
+ <dim>-1</dim>
661
+ </port>
662
+ </output>
663
+ </layer>
664
+ <layer id="46" name="Convert_116" type="Convert" version="opset1">
665
+ <data destination_type="i32" />
666
+ <input>
667
+ <port id="0" precision="BOOL">
668
+ <dim>-1</dim>
669
+ <dim>-1</dim>
670
+ </port>
671
+ </input>
672
+ <output>
673
+ <port id="1" precision="I32">
674
+ <dim>-1</dim>
675
+ <dim>-1</dim>
676
+ </port>
677
+ </output>
678
+ </layer>
679
+ <layer id="47" name="Convert_116.0" type="Convert" version="opset1">
680
+ <data destination_type="i64" />
681
+ <input>
682
+ <port id="0" precision="I32">
683
+ <dim>-1</dim>
684
+ <dim>-1</dim>
685
+ </port>
686
+ </input>
687
+ <output>
688
+ <port id="1" precision="I64" names="attention_mask">
689
+ <dim>-1</dim>
690
+ <dim>-1</dim>
691
+ </port>
692
+ </output>
693
+ </layer>
694
+ <layer id="49" name="RaggedToDense_115.0" type="Convert" version="opset1">
695
+ <data destination_type="i64" />
696
+ <input>
697
+ <port id="0" precision="I32">
698
+ <dim>-1</dim>
699
+ <dim>-1</dim>
700
+ </port>
701
+ </input>
702
+ <output>
703
+ <port id="1" precision="I64" names="input_ids">
704
+ <dim>-1</dim>
705
+ <dim>-1</dim>
706
+ </port>
707
+ </output>
708
+ </layer>
709
+ <layer id="50" name="Result_119" type="Result" version="opset1">
710
+ <input>
711
+ <port id="0" precision="I64">
712
+ <dim>-1</dim>
713
+ <dim>-1</dim>
714
+ </port>
715
+ </input>
716
+ </layer>
717
+ <layer id="48" name="Result_121" type="Result" version="opset1">
718
+ <input>
719
+ <port id="0" precision="I64">
720
+ <dim>-1</dim>
721
+ <dim>-1</dim>
722
+ </port>
723
+ </input>
724
+ </layer>
725
+ </layers>
726
+ <edges>
727
+ <edge from-layer="0" from-port="0" to-layer="5" to-port="0" />
728
+ <edge from-layer="1" from-port="0" to-layer="40" to-port="0" />
729
+ <edge from-layer="2" from-port="0" to-layer="40" to-port="1" />
730
+ <edge from-layer="3" from-port="0" to-layer="40" to-port="2" />
731
+ <edge from-layer="4" from-port="0" to-layer="11" to-port="0" />
732
+ <edge from-layer="5" from-port="1" to-layer="6" to-port="0" />
733
+ <edge from-layer="5" from-port="3" to-layer="18" to-port="4" />
734
+ <edge from-layer="5" from-port="2" to-layer="18" to-port="3" />
735
+ <edge from-layer="5" from-port="1" to-layer="18" to-port="2" />
736
+ <edge from-layer="6" from-port="1" to-layer="9" to-port="0" />
737
+ <edge from-layer="7" from-port="0" to-layer="9" to-port="1" />
738
+ <edge from-layer="8" from-port="0" to-layer="9" to-port="2" />
739
+ <edge from-layer="9" from-port="3" to-layer="14" to-port="0" />
740
+ <edge from-layer="9" from-port="3" to-layer="11" to-port="1" />
741
+ <edge from-layer="10" from-port="0" to-layer="11" to-port="2" />
742
+ <edge from-layer="11" from-port="3" to-layer="18" to-port="0" />
743
+ <edge from-layer="12" from-port="0" to-layer="16" to-port="0" />
744
+ <edge from-layer="13" from-port="0" to-layer="14" to-port="1" />
745
+ <edge from-layer="14" from-port="2" to-layer="16" to-port="1" />
746
+ <edge from-layer="15" from-port="0" to-layer="16" to-port="2" />
747
+ <edge from-layer="16" from-port="3" to-layer="18" to-port="1" />
748
+ <edge from-layer="17" from-port="0" to-layer="18" to-port="5" />
749
+ <edge from-layer="18" from-port="7" to-layer="34" to-port="1" />
750
+ <edge from-layer="18" from-port="6" to-layer="34" to-port="0" />
751
+ <edge from-layer="18" from-port="10" to-layer="21" to-port="2" />
752
+ <edge from-layer="18" from-port="8" to-layer="21" to-port="0" />
753
+ <edge from-layer="18" from-port="9" to-layer="21" to-port="1" />
754
+ <edge from-layer="18" from-port="11" to-layer="21" to-port="3" />
755
+ <edge from-layer="19" from-port="0" to-layer="21" to-port="4" />
756
+ <edge from-layer="20" from-port="0" to-layer="21" to-port="5" />
757
+ <edge from-layer="21" from-port="7" to-layer="24" to-port="1" />
758
+ <edge from-layer="21" from-port="9" to-layer="24" to-port="3" />
759
+ <edge from-layer="21" from-port="8" to-layer="24" to-port="2" />
760
+ <edge from-layer="21" from-port="6" to-layer="24" to-port="0" />
761
+ <edge from-layer="22" from-port="0" to-layer="24" to-port="4" />
762
+ <edge from-layer="23" from-port="0" to-layer="24" to-port="5" />
763
+ <edge from-layer="24" from-port="6" to-layer="34" to-port="2" />
764
+ <edge from-layer="24" from-port="7" to-layer="34" to-port="3" />
765
+ <edge from-layer="24" from-port="8" to-layer="34" to-port="4" />
766
+ <edge from-layer="25" from-port="0" to-layer="26" to-port="0" />
767
+ <edge from-layer="26" from-port="1" to-layer="34" to-port="5" />
768
+ <edge from-layer="26" from-port="3" to-layer="34" to-port="7" />
769
+ <edge from-layer="26" from-port="2" to-layer="34" to-port="6" />
770
+ <edge from-layer="27" from-port="0" to-layer="28" to-port="0" />
771
+ <edge from-layer="28" from-port="1" to-layer="34" to-port="8" />
772
+ <edge from-layer="28" from-port="3" to-layer="34" to-port="10" />
773
+ <edge from-layer="28" from-port="2" to-layer="34" to-port="9" />
774
+ <edge from-layer="29" from-port="0" to-layer="30" to-port="0" />
775
+ <edge from-layer="30" from-port="1" to-layer="34" to-port="11" />
776
+ <edge from-layer="30" from-port="2" to-layer="34" to-port="12" />
777
+ <edge from-layer="30" from-port="3" to-layer="34" to-port="13" />
778
+ <edge from-layer="31" from-port="0" to-layer="32" to-port="0" />
779
+ <edge from-layer="32" from-port="1" to-layer="34" to-port="14" />
780
+ <edge from-layer="32" from-port="2" to-layer="34" to-port="15" />
781
+ <edge from-layer="32" from-port="3" to-layer="34" to-port="16" />
782
+ <edge from-layer="33" from-port="0" to-layer="34" to-port="17" />
783
+ <edge from-layer="34" from-port="18" to-layer="40" to-port="3" />
784
+ <edge from-layer="34" from-port="20" to-layer="40" to-port="5" />
785
+ <edge from-layer="34" from-port="18" to-layer="38" to-port="0" />
786
+ <edge from-layer="34" from-port="18" to-layer="35" to-port="1" />
787
+ <edge from-layer="34" from-port="19" to-layer="35" to-port="0" />
788
+ <edge from-layer="35" from-port="2" to-layer="37" to-port="0" />
789
+ <edge from-layer="36" from-port="0" to-layer="37" to-port="1" />
790
+ <edge from-layer="37" from-port="2" to-layer="38" to-port="1" />
791
+ <edge from-layer="38" from-port="2" to-layer="40" to-port="4" />
792
+ <edge from-layer="39" from-port="0" to-layer="40" to-port="6" />
793
+ <edge from-layer="40" from-port="8" to-layer="41" to-port="0" />
794
+ <edge from-layer="40" from-port="7" to-layer="41" to-port="1" />
795
+ <edge from-layer="40" from-port="7" to-layer="45" to-port="0" />
796
+ <edge from-layer="40" from-port="8" to-layer="45" to-port="1" />
797
+ <edge from-layer="40" from-port="9" to-layer="45" to-port="2" />
798
+ <edge from-layer="41" from-port="2" to-layer="43" to-port="0" />
799
+ <edge from-layer="42" from-port="0" to-layer="43" to-port="1" />
800
+ <edge from-layer="43" from-port="2" to-layer="45" to-port="3" />
801
+ <edge from-layer="44" from-port="0" to-layer="45" to-port="4" />
802
+ <edge from-layer="45" from-port="6" to-layer="46" to-port="0" />
803
+ <edge from-layer="45" from-port="5" to-layer="49" to-port="0" />
804
+ <edge from-layer="46" from-port="1" to-layer="47" to-port="0" />
805
+ <edge from-layer="47" from-port="1" to-layer="48" to-port="0" />
806
+ <edge from-layer="49" from-port="1" to-layer="50" to-port="0" />
807
+ </edges>
808
+ <rt_info>
809
+ <add_attention_mask value="True" />
810
+ <add_prefix_space />
811
+ <add_special_tokens value="True" />
812
+ <bos_token_id value="1" />
813
+ <chat_template value="{% for message in messages %}{{'&lt;|' + message['role'] + '|>' + '&#10;' + message['content'] + '&lt;|end|>&#10;' }}{% endfor %}{% if add_generation_prompt and messages[-1]['role'] != 'assistant' %}{{- '&lt;|assistant|>&#10;' -}}{% endif %}" />
814
+ <clean_up_tokenization_spaces />
815
+ <detokenizer_input_type value="i64" />
816
+ <eos_token_id value="32000" />
817
+ <handle_special_tokens_with_re value="False" />
818
+ <number_of_inputs value="1" />
819
+ <openvino_tokenizers_version value="2025.0.0.0rc2" />
820
+ <openvino_version value="2025.0.0rc2" />
821
+ <original_tokenizer_class value="&lt;class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>" />
822
+ <pad_token_id value="32000" />
823
+ <sentencepiece_version value="0.2.0" />
824
+ <skip_special_tokens value="True" />
825
+ <streaming_detokenizer value="False" />
826
+ <tiktoken_version value="0.8.0" />
827
+ <tokenizer_output_type value="i64" />
828
+ <tokenizers_version value="0.20.1" />
829
+ <transformers_version value="4.45.0" />
830
+ <use_max_padding value="False" />
831
+ <use_sentencepiece_backend value="True" />
832
+ <utf8_replace_mode value="replace" />
833
+ <with_detokenizer value="True" />
834
+ </rt_info>
835
+ </net>
openvino_vision_embeddings_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2da549a09d37697d71f9bc0e238ad041bc471edaa5a879548d90dc0431d4c89b
3
+ size 581818510
openvino_vision_embeddings_model.xml ADDED
The diff for this file is too large to render. See raw diff
 
openvino_vision_projection_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba191044550da50168cd99e4ebc43d0328de09f79ad6aac194914c337cbed742
3
+ size 44052480
openvino_vision_projection_model.xml ADDED
@@ -0,0 +1,270 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <?xml version="1.0"?>
2
+ <net name="Model3" version="11">
3
+ <layers>
4
+ <layer id="0" name="input" type="Parameter" version="opset1">
5
+ <data shape="?,?,4096" element_type="f32" />
6
+ <output>
7
+ <port id="0" precision="FP32" names="input">
8
+ <dim>-1</dim>
9
+ <dim>-1</dim>
10
+ <dim>4096</dim>
11
+ </port>
12
+ </output>
13
+ </layer>
14
+ <layer id="1" name="self.0.weight_compressed" type="Const" version="opset1">
15
+ <data element_type="f16" shape="3072, 4096" offset="0" size="25165824" />
16
+ <output>
17
+ <port id="0" precision="FP16" names="self.0.weight">
18
+ <dim>3072</dim>
19
+ <dim>4096</dim>
20
+ </port>
21
+ </output>
22
+ </layer>
23
+ <layer id="2" name="self.0.weight" type="Convert" version="opset1">
24
+ <data destination_type="f32" />
25
+ <rt_info>
26
+ <attribute name="decompression" version="0" />
27
+ </rt_info>
28
+ <input>
29
+ <port id="0" precision="FP16">
30
+ <dim>3072</dim>
31
+ <dim>4096</dim>
32
+ </port>
33
+ </input>
34
+ <output>
35
+ <port id="1" precision="FP32">
36
+ <dim>3072</dim>
37
+ <dim>4096</dim>
38
+ </port>
39
+ </output>
40
+ </layer>
41
+ <layer id="3" name="__module.0/aten::linear/MatMul" type="MatMul" version="opset1">
42
+ <data transpose_a="false" transpose_b="true" />
43
+ <input>
44
+ <port id="0" precision="FP32">
45
+ <dim>-1</dim>
46
+ <dim>-1</dim>
47
+ <dim>4096</dim>
48
+ </port>
49
+ <port id="1" precision="FP32">
50
+ <dim>3072</dim>
51
+ <dim>4096</dim>
52
+ </port>
53
+ </input>
54
+ <output>
55
+ <port id="2" precision="FP32">
56
+ <dim>-1</dim>
57
+ <dim>-1</dim>
58
+ <dim>3072</dim>
59
+ </port>
60
+ </output>
61
+ </layer>
62
+ <layer id="4" name="Constant_113746_compressed" type="Const" version="opset1">
63
+ <data element_type="f16" shape="1, 1, 3072" offset="25165824" size="6144" />
64
+ <output>
65
+ <port id="0" precision="FP16">
66
+ <dim>1</dim>
67
+ <dim>1</dim>
68
+ <dim>3072</dim>
69
+ </port>
70
+ </output>
71
+ </layer>
72
+ <layer id="5" name="Constant_113746" type="Convert" version="opset1">
73
+ <data destination_type="f32" />
74
+ <rt_info>
75
+ <attribute name="decompression" version="0" />
76
+ </rt_info>
77
+ <input>
78
+ <port id="0" precision="FP16">
79
+ <dim>1</dim>
80
+ <dim>1</dim>
81
+ <dim>3072</dim>
82
+ </port>
83
+ </input>
84
+ <output>
85
+ <port id="1" precision="FP32">
86
+ <dim>1</dim>
87
+ <dim>1</dim>
88
+ <dim>3072</dim>
89
+ </port>
90
+ </output>
91
+ </layer>
92
+ <layer id="6" name="__module.0/aten::linear/Add" type="Add" version="opset1">
93
+ <data auto_broadcast="numpy" />
94
+ <input>
95
+ <port id="0" precision="FP32">
96
+ <dim>-1</dim>
97
+ <dim>-1</dim>
98
+ <dim>3072</dim>
99
+ </port>
100
+ <port id="1" precision="FP32">
101
+ <dim>1</dim>
102
+ <dim>1</dim>
103
+ <dim>3072</dim>
104
+ </port>
105
+ </input>
106
+ <output>
107
+ <port id="2" precision="FP32" names="11">
108
+ <dim>-1</dim>
109
+ <dim>-1</dim>
110
+ <dim>3072</dim>
111
+ </port>
112
+ </output>
113
+ </layer>
114
+ <layer id="7" name="__module.1/aten::gelu/Gelu" type="Gelu" version="opset7">
115
+ <data approximation_mode="ERF" />
116
+ <input>
117
+ <port id="0" precision="FP32">
118
+ <dim>-1</dim>
119
+ <dim>-1</dim>
120
+ <dim>3072</dim>
121
+ </port>
122
+ </input>
123
+ <output>
124
+ <port id="1" precision="FP32" names="13">
125
+ <dim>-1</dim>
126
+ <dim>-1</dim>
127
+ <dim>3072</dim>
128
+ </port>
129
+ </output>
130
+ </layer>
131
+ <layer id="8" name="self.2.weight_compressed" type="Const" version="opset1">
132
+ <data element_type="f16" shape="3072, 3072" offset="25171968" size="18874368" />
133
+ <output>
134
+ <port id="0" precision="FP16" names="self.2.weight">
135
+ <dim>3072</dim>
136
+ <dim>3072</dim>
137
+ </port>
138
+ </output>
139
+ </layer>
140
+ <layer id="9" name="self.2.weight" type="Convert" version="opset1">
141
+ <data destination_type="f32" />
142
+ <rt_info>
143
+ <attribute name="decompression" version="0" />
144
+ </rt_info>
145
+ <input>
146
+ <port id="0" precision="FP16">
147
+ <dim>3072</dim>
148
+ <dim>3072</dim>
149
+ </port>
150
+ </input>
151
+ <output>
152
+ <port id="1" precision="FP32">
153
+ <dim>3072</dim>
154
+ <dim>3072</dim>
155
+ </port>
156
+ </output>
157
+ </layer>
158
+ <layer id="10" name="__module.2/aten::linear/MatMul" type="MatMul" version="opset1">
159
+ <data transpose_a="false" transpose_b="true" />
160
+ <input>
161
+ <port id="0" precision="FP32">
162
+ <dim>-1</dim>
163
+ <dim>-1</dim>
164
+ <dim>3072</dim>
165
+ </port>
166
+ <port id="1" precision="FP32">
167
+ <dim>3072</dim>
168
+ <dim>3072</dim>
169
+ </port>
170
+ </input>
171
+ <output>
172
+ <port id="2" precision="FP32">
173
+ <dim>-1</dim>
174
+ <dim>-1</dim>
175
+ <dim>3072</dim>
176
+ </port>
177
+ </output>
178
+ </layer>
179
+ <layer id="11" name="Constant_113747_compressed" type="Const" version="opset1">
180
+ <data element_type="f16" shape="1, 1, 3072" offset="44046336" size="6144" />
181
+ <output>
182
+ <port id="0" precision="FP16">
183
+ <dim>1</dim>
184
+ <dim>1</dim>
185
+ <dim>3072</dim>
186
+ </port>
187
+ </output>
188
+ </layer>
189
+ <layer id="12" name="Constant_113747" type="Convert" version="opset1">
190
+ <data destination_type="f32" />
191
+ <rt_info>
192
+ <attribute name="decompression" version="0" />
193
+ </rt_info>
194
+ <input>
195
+ <port id="0" precision="FP16">
196
+ <dim>1</dim>
197
+ <dim>1</dim>
198
+ <dim>3072</dim>
199
+ </port>
200
+ </input>
201
+ <output>
202
+ <port id="1" precision="FP32">
203
+ <dim>1</dim>
204
+ <dim>1</dim>
205
+ <dim>3072</dim>
206
+ </port>
207
+ </output>
208
+ </layer>
209
+ <layer id="13" name="__module.2/aten::linear/Add" type="Add" version="opset1">
210
+ <data auto_broadcast="numpy" />
211
+ <input>
212
+ <port id="0" precision="FP32">
213
+ <dim>-1</dim>
214
+ <dim>-1</dim>
215
+ <dim>3072</dim>
216
+ </port>
217
+ <port id="1" precision="FP32">
218
+ <dim>1</dim>
219
+ <dim>1</dim>
220
+ <dim>3072</dim>
221
+ </port>
222
+ </input>
223
+ <output>
224
+ <port id="2" precision="FP32" names="last_hidden_state">
225
+ <dim>-1</dim>
226
+ <dim>-1</dim>
227
+ <dim>3072</dim>
228
+ </port>
229
+ </output>
230
+ </layer>
231
+ <layer id="14" name="Result_112106" type="Result" version="opset1">
232
+ <input>
233
+ <port id="0" precision="FP32">
234
+ <dim>-1</dim>
235
+ <dim>-1</dim>
236
+ <dim>3072</dim>
237
+ </port>
238
+ </input>
239
+ </layer>
240
+ </layers>
241
+ <edges>
242
+ <edge from-layer="0" from-port="0" to-layer="3" to-port="0" />
243
+ <edge from-layer="1" from-port="0" to-layer="2" to-port="0" />
244
+ <edge from-layer="2" from-port="1" to-layer="3" to-port="1" />
245
+ <edge from-layer="3" from-port="2" to-layer="6" to-port="0" />
246
+ <edge from-layer="4" from-port="0" to-layer="5" to-port="0" />
247
+ <edge from-layer="5" from-port="1" to-layer="6" to-port="1" />
248
+ <edge from-layer="6" from-port="2" to-layer="7" to-port="0" />
249
+ <edge from-layer="7" from-port="1" to-layer="10" to-port="0" />
250
+ <edge from-layer="8" from-port="0" to-layer="9" to-port="0" />
251
+ <edge from-layer="9" from-port="1" to-layer="10" to-port="1" />
252
+ <edge from-layer="10" from-port="2" to-layer="13" to-port="0" />
253
+ <edge from-layer="11" from-port="0" to-layer="12" to-port="0" />
254
+ <edge from-layer="12" from-port="1" to-layer="13" to-port="1" />
255
+ <edge from-layer="13" from-port="2" to-layer="14" to-port="0" />
256
+ </edges>
257
+ <rt_info>
258
+ <Runtime_version value="2025.0.0-17933-815af98acd8-releases/2025/0" />
259
+ <conversion_parameters>
260
+ <framework value="pytorch" />
261
+ <is_python_object value="True" />
262
+ </conversion_parameters>
263
+ <optimum>
264
+ <optimum_intel_version value="1.22.0.dev0+753f84d" />
265
+ <optimum_version value="1.24.0.dev0" />
266
+ <pytorch_version value="2.5.0+cpu" />
267
+ <transformers_version value="4.45.0" />
268
+ </optimum>
269
+ </rt_info>
270
+ </net>
preprocessor_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_map": {
3
+ "AutoImageProcessor": "microsoft/Phi-3.5-vision-instruct--processing_phi3_v.Phi3VImageProcessor",
4
+ "AutoProcessor": "microsoft/Phi-3.5-vision-instruct--processing_phi3_v.Phi3VProcessor"
5
+ },
6
+ "do_convert_rgb": true,
7
+ "image_mean": [
8
+ 0.48145466,
9
+ 0.4578275,
10
+ 0.40821073
11
+ ],
12
+ "image_processor_type": "Phi3VImageProcessor",
13
+ "image_std": [
14
+ 0.26862954,
15
+ 0.26130258,
16
+ 0.27577711
17
+ ],
18
+ "num_crops": 4,
19
+ "num_img_tokens": 144,
20
+ "processor_class": "Phi3VProcessor"
21
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|system|>",
4
+ "<|end|>",
5
+ "<|user|>",
6
+ "<|end|>"
7
+ ],
8
+ "bos_token": {
9
+ "content": "<s>",
10
+ "lstrip": false,
11
+ "normalized": false,
12
+ "rstrip": false,
13
+ "single_word": false
14
+ },
15
+ "eos_token": {
16
+ "content": "<|endoftext|>",
17
+ "lstrip": false,
18
+ "normalized": false,
19
+ "rstrip": false,
20
+ "single_word": false
21
+ },
22
+ "pad_token": {
23
+ "content": "<|endoftext|>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false
28
+ },
29
+ "unk_token": {
30
+ "content": "<unk>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false
35
+ }
36
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,413 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": true,
27
+ "single_word": false,
28
+ "special": false
29
+ },
30
+ "32000": {
31
+ "content": "<|endoftext|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "32001": {
39
+ "content": "<|assistant|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": true,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "32002": {
47
+ "content": "<|placeholder1|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": true,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "32003": {
55
+ "content": "<|placeholder2|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": true,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "32004": {
63
+ "content": "<|placeholder3|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": true,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "32005": {
71
+ "content": "<|placeholder4|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": true,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "32006": {
79
+ "content": "<|system|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "32007": {
87
+ "content": "<|end|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "32008": {
95
+ "content": "<|placeholder5|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": true,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "32009": {
103
+ "content": "<|placeholder6|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": true,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "32010": {
111
+ "content": "<|user|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "32011": {
119
+ "content": "<|placeholder7|>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": true,
123
+ "single_word": false,
124
+ "special": true
125
+ },
126
+ "32012": {
127
+ "content": "<|placeholder8|>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": true,
131
+ "single_word": false,
132
+ "special": true
133
+ },
134
+ "32013": {
135
+ "content": "<|placeholder9|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": true,
139
+ "single_word": false,
140
+ "special": true
141
+ },
142
+ "32014": {
143
+ "content": "<|placeholder10|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": true,
147
+ "single_word": false,
148
+ "special": true
149
+ },
150
+ "32015": {
151
+ "content": "<|placeholder11|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": true,
155
+ "single_word": false,
156
+ "special": true
157
+ },
158
+ "32016": {
159
+ "content": "<|placeholder12|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": true,
163
+ "single_word": false,
164
+ "special": true
165
+ },
166
+ "32017": {
167
+ "content": "<|placeholder13|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": true,
171
+ "single_word": false,
172
+ "special": true
173
+ },
174
+ "32018": {
175
+ "content": "<|placeholder14|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": true,
179
+ "single_word": false,
180
+ "special": true
181
+ },
182
+ "32019": {
183
+ "content": "<|placeholder15|>",
184
+ "lstrip": false,
185
+ "normalized": false,
186
+ "rstrip": true,
187
+ "single_word": false,
188
+ "special": true
189
+ },
190
+ "32020": {
191
+ "content": "<|placeholder16|>",
192
+ "lstrip": false,
193
+ "normalized": false,
194
+ "rstrip": true,
195
+ "single_word": false,
196
+ "special": true
197
+ },
198
+ "32021": {
199
+ "content": "<|placeholder17|>",
200
+ "lstrip": false,
201
+ "normalized": false,
202
+ "rstrip": true,
203
+ "single_word": false,
204
+ "special": true
205
+ },
206
+ "32022": {
207
+ "content": "<|placeholder18|>",
208
+ "lstrip": false,
209
+ "normalized": false,
210
+ "rstrip": true,
211
+ "single_word": false,
212
+ "special": true
213
+ },
214
+ "32023": {
215
+ "content": "<|placeholder19|>",
216
+ "lstrip": false,
217
+ "normalized": false,
218
+ "rstrip": true,
219
+ "single_word": false,
220
+ "special": true
221
+ },
222
+ "32024": {
223
+ "content": "<|placeholder20|>",
224
+ "lstrip": false,
225
+ "normalized": false,
226
+ "rstrip": true,
227
+ "single_word": false,
228
+ "special": true
229
+ },
230
+ "32025": {
231
+ "content": "<|placeholder21|>",
232
+ "lstrip": false,
233
+ "normalized": false,
234
+ "rstrip": true,
235
+ "single_word": false,
236
+ "special": true
237
+ },
238
+ "32026": {
239
+ "content": "<|placeholder22|>",
240
+ "lstrip": false,
241
+ "normalized": false,
242
+ "rstrip": true,
243
+ "single_word": false,
244
+ "special": true
245
+ },
246
+ "32027": {
247
+ "content": "<|placeholder23|>",
248
+ "lstrip": false,
249
+ "normalized": false,
250
+ "rstrip": true,
251
+ "single_word": false,
252
+ "special": true
253
+ },
254
+ "32028": {
255
+ "content": "<|placeholder24|>",
256
+ "lstrip": false,
257
+ "normalized": false,
258
+ "rstrip": true,
259
+ "single_word": false,
260
+ "special": true
261
+ },
262
+ "32029": {
263
+ "content": "<|placeholder25|>",
264
+ "lstrip": false,
265
+ "normalized": false,
266
+ "rstrip": true,
267
+ "single_word": false,
268
+ "special": true
269
+ },
270
+ "32030": {
271
+ "content": "<|placeholder26|>",
272
+ "lstrip": false,
273
+ "normalized": false,
274
+ "rstrip": true,
275
+ "single_word": false,
276
+ "special": true
277
+ },
278
+ "32031": {
279
+ "content": "<|placeholder27|>",
280
+ "lstrip": false,
281
+ "normalized": false,
282
+ "rstrip": true,
283
+ "single_word": false,
284
+ "special": true
285
+ },
286
+ "32032": {
287
+ "content": "<|placeholder28|>",
288
+ "lstrip": false,
289
+ "normalized": false,
290
+ "rstrip": true,
291
+ "single_word": false,
292
+ "special": true
293
+ },
294
+ "32033": {
295
+ "content": "<|placeholder29|>",
296
+ "lstrip": false,
297
+ "normalized": false,
298
+ "rstrip": true,
299
+ "single_word": false,
300
+ "special": true
301
+ },
302
+ "32034": {
303
+ "content": "<|placeholder30|>",
304
+ "lstrip": false,
305
+ "normalized": false,
306
+ "rstrip": true,
307
+ "single_word": false,
308
+ "special": true
309
+ },
310
+ "32035": {
311
+ "content": "<|placeholder31|>",
312
+ "lstrip": false,
313
+ "normalized": false,
314
+ "rstrip": true,
315
+ "single_word": false,
316
+ "special": true
317
+ },
318
+ "32036": {
319
+ "content": "<|placeholder32|>",
320
+ "lstrip": false,
321
+ "normalized": false,
322
+ "rstrip": true,
323
+ "single_word": false,
324
+ "special": true
325
+ },
326
+ "32037": {
327
+ "content": "<|placeholder33|>",
328
+ "lstrip": false,
329
+ "normalized": false,
330
+ "rstrip": true,
331
+ "single_word": false,
332
+ "special": true
333
+ },
334
+ "32038": {
335
+ "content": "<|placeholder34|>",
336
+ "lstrip": false,
337
+ "normalized": false,
338
+ "rstrip": true,
339
+ "single_word": false,
340
+ "special": true
341
+ },
342
+ "32039": {
343
+ "content": "<|placeholder35|>",
344
+ "lstrip": false,
345
+ "normalized": false,
346
+ "rstrip": true,
347
+ "single_word": false,
348
+ "special": true
349
+ },
350
+ "32040": {
351
+ "content": "<|placeholder36|>",
352
+ "lstrip": false,
353
+ "normalized": false,
354
+ "rstrip": true,
355
+ "single_word": false,
356
+ "special": true
357
+ },
358
+ "32041": {
359
+ "content": "<|placeholder37|>",
360
+ "lstrip": false,
361
+ "normalized": false,
362
+ "rstrip": true,
363
+ "single_word": false,
364
+ "special": true
365
+ },
366
+ "32042": {
367
+ "content": "<|placeholder38|>",
368
+ "lstrip": false,
369
+ "normalized": false,
370
+ "rstrip": true,
371
+ "single_word": false,
372
+ "special": true
373
+ },
374
+ "32043": {
375
+ "content": "<|placeholder39|>",
376
+ "lstrip": false,
377
+ "normalized": false,
378
+ "rstrip": true,
379
+ "single_word": false,
380
+ "special": true
381
+ },
382
+ "32044": {
383
+ "content": "<|image|>",
384
+ "lstrip": false,
385
+ "normalized": false,
386
+ "rstrip": true,
387
+ "single_word": false,
388
+ "special": true
389
+ }
390
+ },
391
+ "additional_special_tokens": [
392
+ "<|system|>",
393
+ "<|end|>",
394
+ "<|user|>",
395
+ "<|end|>"
396
+ ],
397
+ "auto_map": {
398
+ "AutoProcessor": "microsoft/Phi-3.5-vision-instruct--processing_phi3_v.Phi3VProcessor"
399
+ },
400
+ "bos_token": "<s>",
401
+ "chat_template": "{% for message in messages %}{{'<|' + message['role'] + '|>' + '\n' + message['content'] + '<|end|>\n' }}{% endfor %}{% if add_generation_prompt and messages[-1]['role'] != 'assistant' %}{{- '<|assistant|>\n' -}}{% endif %}",
402
+ "clean_up_tokenization_spaces": false,
403
+ "eos_token": "<|endoftext|>",
404
+ "legacy": false,
405
+ "model_max_length": 131072,
406
+ "pad_token": "<|endoftext|>",
407
+ "padding_side": "right",
408
+ "processor_class": "Phi3VProcessor",
409
+ "sp_model_kwargs": {},
410
+ "tokenizer_class": "LlamaTokenizer",
411
+ "unk_token": "<unk>",
412
+ "use_default_system_prompt": false
413
+ }