Hugging Face's logo

addyo07
/
vox-models

ONNX
GGUF
generic
voice
stt
tts
llm
vox
real-time
edge-inference
imatrix
conversational
Model card Files
xet
 Community
addy-hypr4 commited on
Commit
744f970
·
1 Parent(s): d92eadb

feat(models): update active modernbert memory scope, edge creation, and deberta v3 base nli models

Browse files
Files changed (13) hide show
  1. .gitattributes +3 -1
  2. classifier/{distilbert-query-classifier → modernbert_edge_creation}/model_quantized.onnx +2 -2
  3. classifier/{distilbert-query-classifier → modernbert_edge_creation}/tokenizer.json +0 -0
  4. classifier/modernbert_memory_scope/model_quantized.onnx +3 -0
  5. classifier/modernbert_memory_scope/tokenizer.json +0 -0
  6. classifier/modernbert_memory_scope/tokenizer_config.json +17 -0
  7. models_manifest.json +0 -370
  8. models_manifest.json +1 -0
  9. nli/nli-deberta-v3-base/README.md +3 -0
  10. nli/nli-deberta-v3-base/added_tokens.json +3 -0
  11. nli/nli-deberta-v3-base/config.json +3 -0
  12. nli/nli-deberta-v3-base/model_quantized.onnx +3 -0
  13. nli/nli-deberta-v3-base/tokenizer.json +3 -0
.gitattributes CHANGED
@@ -56,4 +56,6 @@ tts/chatterbox/ filter=lfs diff=lfs merge=lfs -text
56
  embedding/**/* filter=lfs diff=lfs merge=lfs -text
57
  nli/**/* filter=lfs diff=lfs merge=lfs -text
58
  *.wav filter=lfs diff=lfs merge=lfs -text
59
-
 
 
 
56
  embedding/**/* filter=lfs diff=lfs merge=lfs -text
57
  nli/**/* filter=lfs diff=lfs merge=lfs -text
58
  *.wav filter=lfs diff=lfs merge=lfs -text
59
+ classifier/modernbert_edge_creation filter=lfs diff=lfs merge=lfs -text
60
+ classifier/modernbert_memory_scope filter=lfs diff=lfs merge=lfs -text
61
+ nli/nli-deberta-v3-base filter=lfs diff=lfs merge=lfs -text
classifier/{distilbert-query-classifier → modernbert_edge_creation}/model_quantized.onnx RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5093f8d2e2252cd952e134231fb784809bc45598d278f89e56c8a320017fdeae
3
- size 135740553
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4118dd78d49db25f2d6b44a7f23b31886222f50c85a3f01134578cee5e27f4f7
3
+ size 150653243
classifier/{distilbert-query-classifier → modernbert_edge_creation}/tokenizer.json RENAMED
The diff for this file is too large to render. See raw diff
 
classifier/modernbert_memory_scope/model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a24244f282966ded439a9e1c669d739b3fcbdebd52adeada642b4d364e9ffd8
3
+ size 150647789
classifier/modernbert_memory_scope/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
classifier/modernbert_memory_scope/tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "clean_up_tokenization_spaces": true,
4
+ "cls_token": "[CLS]",
5
+ "is_local": false,
6
+ "local_files_only": false,
7
+ "mask_token": "[MASK]",
8
+ "model_input_names": [
9
+ "input_ids",
10
+ "attention_mask"
11
+ ],
12
+ "model_max_length": 8192,
13
+ "pad_token": "[PAD]",
14
+ "sep_token": "[SEP]",
15
+ "tokenizer_class": "TokenizersBackend",
16
+ "unk_token": "[UNK]"
17
+ }
models_manifest.json DELETED
@@ -1,370 +0,0 @@
1
- {
2
- "models_version": "1.3.0",
3
- "release_notes": [
4
- "Standardized canonical model directory structure and file names across all engines.",
5
- "MiniLM-L12-v2 integrated as primary multilingual embedding engine.",
6
- "BGE-M3 configured as fallback embedding engine.",
7
- "DistilBERT Query Classifier integrated as primary memory intent routing engine.",
8
- "Nemotron 3.5, Qwen3 ASR, DeBERTa-v3 NLI, Supertonic 3, and Chatterbox TTS paths standardized."
9
- ],
10
- "total_size_bytes": 10219831741,
11
- "model_groups": [
12
- {
13
- "id": "ten_vad",
14
- "name": "Ten VAD Engine",
15
- "category": "vad",
16
- "version": "1.0.0",
17
- "files": [
18
- {
19
- "id": "ten_vad_onnx",
20
- "path": "vad/ten_vad.onnx",
21
- "size": 332211,
22
- "sha256": "718cb7eef47e3cf5ddbe7e967a7503f46b8b469c0706872f494dfa921b486206",
23
- "required": true
24
- }
25
- ]
26
- },
27
- {
28
- "id": "vox_translit_rnn",
29
- "name": "Vox Transliteration Engine",
30
- "category": "translit",
31
- "version": "1.0.0",
32
- "files": [
33
- {
34
- "id": "translit_encoder",
35
- "path": "translit/encoder.onnx",
36
- "size": 9529392,
37
- "sha256": "af1ed31d5af319a353587c746b5c7540646e154d977d7614081c30f8bfb40564",
38
- "required": true
39
- },
40
- {
41
- "id": "translit_decoder",
42
- "path": "translit/decoder.onnx",
43
- "size": 6819685,
44
- "sha256": "4930f3460baf4c0f75a346e1320579469b4ff5e00d3e62375cbcb0cee5120aa3",
45
- "required": true
46
- },
47
- {
48
- "id": "translit_input_vocab",
49
- "path": "translit/input_vocab.json",
50
- "size": 1395,
51
- "sha256": "b7126832254c449c9a208ec650feae5682a7f77323fd2e9bf2ad29c1ab7332e2",
52
- "required": true
53
- },
54
- {
55
- "id": "translit_target_vocab",
56
- "path": "translit/target_vocab.json",
57
- "size": 787,
58
- "sha256": "bd5e364110187e05fcbf00529af0facf45a7838e2aa6b410390af1d80ff08f96",
59
- "required": true
60
- }
61
- ]
62
- },
63
- {
64
- "id": "nvidia_nemotron",
65
- "name": "Nemotron-3.5 ASR Engine",
66
- "category": "stt",
67
- "version": "1.0.0",
68
- "files": [
69
- {
70
- "id": "stt_nemotron_config",
71
- "path": "stt/nemotron-3.5/config.json",
72
- "size": 602,
73
- "sha256": "0592a224f30670e00f5e0f1f106f0171428b0abde0fd4cbc64237b02c2241502",
74
- "required": true
75
- },
76
- {
77
- "id": "stt_nemotron_decoder_joint",
78
- "path": "stt/nemotron-3.5/decoder_joint.onnx",
79
- "size": 97590054,
80
- "sha256": "634dfadf24cb4f73c2fae170b36611d68db48186426882cbc8f7e02ed9f2bb29",
81
- "required": true
82
- },
83
- {
84
- "id": "stt_nemotron_encoder",
85
- "path": "stt/nemotron-3.5/encoder.onnx",
86
- "size": 657572215,
87
- "sha256": "e653d494a4855ee87f6251e8a9677416a73edf6f36578841316fa2b77a53cda7",
88
- "required": true
89
- },
90
- {
91
- "id": "stt_nemotron_tokenizer",
92
- "path": "stt/nemotron-3.5/tokenizer.model",
93
- "size": 406554,
94
- "sha256": "ce3895e40806f02a26c3a225161b96ef682d6c0054bae32a245dec4258d7d291",
95
- "required": true
96
- }
97
- ]
98
- },
99
- {
100
- "id": "qwen3_asr",
101
- "name": "Qwen3 ASR Engine",
102
- "category": "stt",
103
- "version": "1.0.0",
104
- "files": [
105
- {
106
- "id": "stt_conv_frontend",
107
- "path": "stt/qwen3-asr/conv_frontend.onnx",
108
- "size": 44148281,
109
- "sha256": "1f38b879b0871c9f689b092ad4fa85c68ef473c585dcf53d50a029fe8a0ab0cb",
110
- "required": false
111
- },
112
- {
113
- "id": "stt_encoder",
114
- "path": "stt/qwen3-asr/encoder.int8.onnx",
115
- "size": 182491662,
116
- "sha256": "ae0cf814d9cd5249b98a1aa9199511d0c0a0c5c8b3059532637fa3cb4785c60c",
117
- "required": false
118
- },
119
- {
120
- "id": "stt_decoder",
121
- "path": "stt/qwen3-asr/decoder.int8.onnx",
122
- "size": 755913551,
123
- "sha256": "768f0cc54a804e5b9724133910071995ce03ecd1a4d5be8e5e766b022623dc8d",
124
- "required": false
125
- },
126
- {
127
- "id": "stt_vocab",
128
- "path": "stt/qwen3-asr/tokenizer/vocab.json",
129
- "size": 2776833,
130
- "sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
131
- "required": false
132
- },
133
- {
134
- "id": "stt_merges",
135
- "path": "stt/qwen3-asr/tokenizer/merges.txt",
136
- "size": 1671853,
137
- "sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5",
138
- "required": false
139
- },
140
- {
141
- "id": "stt_config",
142
- "path": "stt/qwen3-asr/tokenizer/tokenizer_config.json",
143
- "size": 12487,
144
- "sha256": "4942d005604266809309cabc9f4e9cb89ce855d59b14681fdc0e1cc62ea26c4c",
145
- "required": false
146
- }
147
- ]
148
- },
149
- {
150
- "id": "gemma_4_reasoning",
151
- "name": "Gemma-4 Reasoning LLM",
152
- "category": "llm",
153
- "version": "1.0.0",
154
- "files": [
155
- {
156
- "id": "llm_gemma_4_q4_k_m",
157
- "path": "llm/gemma4/gemma-4-e2b-q4_k_m.gguf",
158
- "size": 3462678272,
159
- "sha256": "b5310340b3a23d31655d7119d100d5df1b2d8ee17b3ca8b0a23ad7e9eb5fa705",
160
- "required": false
161
- }
162
- ]
163
- },
164
- {
165
- "id": "llama_3_2_reasoning_q4",
166
- "name": "Llama-3.2 Reasoning LLM (Q4)",
167
- "category": "llm",
168
- "version": "1.0.0",
169
- "files": [
170
- {
171
- "id": "llm_llama_3_2_1b_instruct_q4_k_m",
172
- "path": "llm/llama/llama-3.2-1b-q4_k_m.gguf",
173
- "size": 807694464,
174
- "sha256": "6f85a640a97cf2bf5b8e764087b1e83da0fdb51d7c9fab7d0fece9385611df83",
175
- "required": false
176
- }
177
- ]
178
- },
179
- {
180
- "id": "llama_3_2_reasoning",
181
- "name": "Llama-3.2 Reasoning LLM (Q6)",
182
- "category": "llm",
183
- "version": "1.0.0",
184
- "files": [
185
- {
186
- "id": "llm_llama_3_2_1b_instruct_q6_k",
187
- "path": "llm/llama/llama-3.2-1b-q6_k.gguf",
188
- "size": 1021800576,
189
- "sha256": "0f4c510daf16e0d1b3bc94931fd9296c28936bebdda2593687d4eb70c5b70628",
190
- "required": false
191
- }
192
- ]
193
- },
194
- {
195
- "id": "supertonic_tts",
196
- "name": "Supertonic 3 TTS Engine",
197
- "category": "tts",
198
- "version": "3.0.0",
199
- "files": [
200
- {
201
- "id": "tts_supertonic_duration_predictor",
202
- "path": "tts/supertonic-3/duration_predictor.int8.onnx",
203
- "size": 3700147,
204
- "sha256": "c3eb91414d5ff8a7a239b7fe9e34e7e2bf8a8140d8375ffb14718b1c639325db",
205
- "required": false
206
- },
207
- {
208
- "id": "tts_supertonic_text_encoder",
209
- "path": "tts/supertonic-3/text_encoder.int8.onnx",
210
- "size": 36416150,
211
- "sha256": "c7befd5ea8c3119769e8a6c1486c4edc6a3bc8365c67621c881bbb774b9902ff",
212
- "required": false
213
- },
214
- {
215
- "id": "tts_supertonic_vector_estimator",
216
- "path": "tts/supertonic-3/vector_estimator.int8.onnx",
217
- "size": 78400833,
218
- "sha256": "20cd86fa5c6effedfda0e7cffe5b0569ca401c440a0c3a1d72bf39286c0db3fd",
219
- "required": false
220
- },
221
- {
222
- "id": "tts_supertonic_vocoder",
223
- "path": "tts/supertonic-3/vocoder.int8.onnx",
224
- "size": 25991073,
225
- "sha256": "e923d60f53f95eb1ce235f1dc33ec56d9c057823c96fa6f8acf98f32b0da6152",
226
- "required": false
227
- },
228
- {
229
- "id": "tts_supertonic_config",
230
- "path": "tts/supertonic-3/tts.json",
231
- "size": 8253,
232
- "sha256": "42078d3aef1cd43ab43021f3c54f47d2d75ceb4e75f627f118890128b06a0d09",
233
- "required": false
234
- },
235
- {
236
- "id": "tts_supertonic_indexer",
237
- "path": "tts/supertonic-3/unicode_indexer.bin",
238
- "size": 262144,
239
- "sha256": "8402ca48e5189a8950138580b0fff64db6f072f24ac07cd54ba8b2fbb9883b30",
240
- "required": false
241
- },
242
- {
243
- "id": "tts_supertonic_voice_bin",
244
- "path": "tts/supertonic-3/voice.bin",
245
- "size": 517168,
246
- "sha256": "67d5209b0ee8ce6c74105ffbe12fe6a7628aea3b4ba2fcb308a4a67938a93ce8",
247
- "required": false
248
- }
249
- ]
250
- },
251
- {
252
- "id": "chatterbox_tts",
253
- "name": "Chatterbox Multilingual TTS Engine",
254
- "category": "tts",
255
- "version": "1.0.0",
256
- "files": [
257
- {
258
- "id": "tts_chatterbox_t3",
259
- "path": "tts/chatterbox/t3-q4_0.gguf",
260
- "size": 344985408,
261
- "sha256": "a7c8cbb267a836697a39757ada25a8ab095712b74961af2b86012dbd296a3a7b",
262
- "required": false
263
- },
264
- {
265
- "id": "tts_chatterbox_s3gen",
266
- "path": "tts/chatterbox/s3gen-f16.gguf",
267
- "size": 1056431360,
268
- "sha256": "b25fe64c5b5287952c82684f6cadb12698257dfe8ed3d9e578801ca970b59664",
269
- "required": false
270
- },
271
- {
272
- "id": "tts_chatterbox_voices",
273
- "path": "tts/chatterbox/voices.tar.gz",
274
- "size": 18897732,
275
- "sha256": "50f998b9837e9b8934e1ad946fe2d32555a472379bb1d2982a4938ed4262fda0",
276
- "archive": "tar.gz",
277
- "required": false
278
- }
279
- ]
280
- },
281
- {
282
- "id": "bge_m3",
283
- "name": "BGE-M3 Multilingual Embedding Engine (Fallback)",
284
- "category": "embedding",
285
- "version": "1.0.0",
286
- "files": [
287
- {
288
- "id": "embedding_bge_m3_onnx",
289
- "path": "embedding/bge-m3/model_quantized.onnx",
290
- "size": 569694530,
291
- "sha256": "0826f8c1ab9edf1801db86c61919d4d108e8bfc0b809ec823ad366882ff0b77d",
292
- "required": false
293
- },
294
- {
295
- "id": "embedding_bge_m3_tokenizer",
296
- "path": "embedding/bge-m3/tokenizer.json",
297
- "size": 17082821,
298
- "sha256": "6710678b12670bc442b99edc952c4d996ae309a7020c1fa0096dd245c2faf790",
299
- "required": false
300
- }
301
- ]
302
- },
303
- {
304
- "id": "minilm_l12_v2",
305
- "name": "MiniLM-L12 Multilingual Embedding Engine (Primary)",
306
- "category": "embedding",
307
- "version": "1.0.0",
308
- "files": [
309
- {
310
- "id": "embedding_minilm_l12_onnx",
311
- "path": "embedding/minilm-l12-v2/model_int8.onnx",
312
- "size": 118054609,
313
- "sha256": "4b76e8b4e72390a880356c38be28b9d363d6642d99dcf5c0c978bb7ecb3df637",
314
- "required": true
315
- },
316
- {
317
- "id": "embedding_minilm_l12_tokenizer",
318
- "path": "embedding/minilm-l12-v2/tokenizer.json",
319
- "size": 17082821,
320
- "sha256": "6710678b12670bc442b99edc952c4d996ae309a7020c1fa0096dd245c2faf790",
321
- "required": true
322
- }
323
- ]
324
- },
325
- {
326
- "id": "deberta_v3_xsmall_nli",
327
- "name": "DeBERTa-v3 XSmall NLI Contradiction Engine",
328
- "category": "nli",
329
- "version": "1.0.0",
330
- "files": [
331
- {
332
- "id": "nli_deberta_onnx",
333
- "path": "nli/deberta-v3-xsmall/model_quantized.onnx",
334
- "size": 87246587,
335
- "sha256": "3fac2500c45c75af42c7711de0d1b93d59577456100208be0dc1f9e8811946b6",
336
- "required": true
337
- },
338
- {
339
- "id": "nli_deberta_tokenizer",
340
- "path": "nli/deberta-v3-xsmall/tokenizer.json",
341
- "size": 8656551,
342
- "sha256": "a86f883318afa11c8c10466f1bf4efaeb6ded28a52cbe57217a8fa0d0a2a87df",
343
- "required": true
344
- }
345
- ]
346
- },
347
- {
348
- "id": "distilbert_query_classifier",
349
- "name": "DistilBERT Query Classifier Engine",
350
- "category": "classifier",
351
- "version": "1.0.0",
352
- "files": [
353
- {
354
- "id": "classifier_model_quantized",
355
- "path": "classifier/distilbert-query-classifier/model_quantized.onnx",
356
- "size": 135740553,
357
- "sha256": "5093f8d2e2252cd952e134231fb784809bc45598d278f89e56c8a320017fdeae",
358
- "required": true
359
- },
360
- {
361
- "id": "classifier_tokenizer",
362
- "path": "classifier/distilbert-query-classifier/tokenizer.json",
363
- "size": 2919625,
364
- "sha256": "519ee3affedf69aac5333eb1963ce157b17c67e650f9a9903c561d2f47cf02bb",
365
- "required": true
366
- }
367
- ]
368
- }
369
- ]
370
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
models_manifest.json ADDED
@@ -0,0 +1 @@
 
 
1
+ /home/addy/projects/apps/vox/manifests/models_manifest.json
nli/nli-deberta-v3-base/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b5f35ce954c5a0cdc2669e4d77ce3a58d2f3874ac908f025e585626c06fd039
3
+ size 1164
nli/nli-deberta-v3-base/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc046d04c9b0ada7ae6f1dc89c465801799acdf0c9a6aab8c15a1b2d5ca4e91f
3
+ size 23
nli/nli-deberta-v3-base/config.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cedce9043e58e52f0f6fc31e8c87ec518092a742851e80eddbb651d36a0c754f
3
+ size 1037
nli/nli-deberta-v3-base/model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d83d9d0ba179cab8b78b942dcc42756da7712b2bbe101a2a4c25fdfa7968e2f
3
+ size 244291931
nli/nli-deberta-v3-base/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a86f883318afa11c8c10466f1bf4efaeb6ded28a52cbe57217a8fa0d0a2a87df
3
+ size 8656551