addy-hypr4 commited on
Commit
b536905
·
1 Parent(s): 2f0c401

Standardize model asset paths and update models manifest

Browse files
embedding/minilm-l12-v2/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1faca2377fb36a7c5ee603ffdd170cf93126be37e7b3b422d319b8ed34081e73
3
+ size 1151
embedding/minilm-l12-v2/config.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05b570bff786faa5c4604152aa16f19f77ed6dfc31e47dd0f3dd987078693ac7
3
+ size 673
embedding/minilm-l12-v2/model_int8.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6ea442ff6a891daefed7c83b2f596fc5dc66bf697e4d006236f64f34bbcf4c8
3
+ size 118054609
embedding/minilm-l12-v2/special_tokens_map.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06e405a36dfe4b9604f484f6a1e619af1a7f7d09e34a8555eb0b77b66318067f
3
+ size 280
embedding/minilm-l12-v2/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b60b6b43406a48bf3638526314f3d232d97058bc93472ff2de930d43686fa441
3
+ size 17082913
embedding/minilm-l12-v2/tokenizer_config.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f5961b9ac86288cccdb97f32fb848d6187c78e1603958c53f3ea1f296b7d8a2
3
+ size 496
embedding/minilm-l12-v2/unigram.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71b44701d7efd054205115acfa6ef126c5d2f84bd3affe0c59e48163674d19a6
3
+ size 14763234
llm/gemma4/{google_gemma-4-E2B-it-Q4_K_M.gguf → gemma-4-e2b-q4_k_m.gguf} RENAMED
File without changes
llm/llama/{Llama-3.2-1B-Instruct-Q4_K_M.gguf → llama-3.2-1b-q4_k_m.gguf} RENAMED
File without changes
llm/llama/{Llama-3.2-1B-Instruct-Q6_K.gguf → llama-3.2-1b-q6_k.gguf} RENAMED
File without changes
models_manifest.json CHANGED
@@ -1,9 +1,10 @@
1
  {
2
- "models_version": "1.1.0",
3
  "release_notes": [
4
- "Portable, single-instance TTS engine with 31 languages and flow-matching architecture.",
5
- "Supertonic 3 replaces Kokoro (EN) + Piper (HI) as the sole TTS engine.",
6
- "Chatterbox TTS integrated"
 
7
  ],
8
  "total_size_bytes": 10081171563,
9
  "model_groups": [
@@ -66,28 +67,28 @@
66
  "files": [
67
  {
68
  "id": "stt_nemotron_config",
69
- "path": "stt/nvidia-nemotron-3.5/config.json",
70
  "size": 602,
71
  "sha256": "0592a224f30670e00f5e0f1f106f0171428b0abde0fd4cbc64237b02c2241502",
72
  "required": true
73
  },
74
  {
75
  "id": "stt_nemotron_decoder_joint",
76
- "path": "stt/nvidia-nemotron-3.5/decoder_joint.onnx",
77
  "size": 97590054,
78
  "sha256": "634dfadf24cb4f73c2fae170b36611d68db48186426882cbc8f7e02ed9f2bb29",
79
  "required": true
80
  },
81
  {
82
  "id": "stt_nemotron_encoder",
83
- "path": "stt/nvidia-nemotron-3.5/encoder.onnx",
84
  "size": 657572215,
85
  "sha256": "e653d494a4855ee87f6251e8a9677416a73edf6f36578841316fa2b77a53cda7",
86
  "required": true
87
  },
88
  {
89
  "id": "stt_nemotron_tokenizer",
90
- "path": "stt/nvidia-nemotron-3.5/tokenizer.model",
91
  "size": 406554,
92
  "sha256": "ce3895e40806f02a26c3a225161b96ef682d6c0054bae32a245dec4258d7d291",
93
  "required": true
@@ -152,7 +153,7 @@
152
  "files": [
153
  {
154
  "id": "llm_gemma_4_q4_k_m",
155
- "path": "llm/gemma4/google_gemma-4-E2B-it-Q4_K_M.gguf",
156
  "size": 3462678272,
157
  "sha256": "b5310340b3a23d31655d7119d100d5df1b2d8ee17b3ca8b0a23ad7e9eb5fa705",
158
  "required": false
@@ -167,7 +168,7 @@
167
  "files": [
168
  {
169
  "id": "llm_llama_3_2_1b_instruct_q4_k_m",
170
- "path": "llm/llama/Llama-3.2-1B-Instruct-Q4_K_M.gguf",
171
  "size": 807694464,
172
  "sha256": "6f85a640a97cf2bf5b8e764087b1e83da0fdb51d7c9fab7d0fece9385611df83",
173
  "required": false
@@ -182,28 +183,13 @@
182
  "files": [
183
  {
184
  "id": "llm_llama_3_2_1b_instruct_q6_k",
185
- "path": "llm/llama/Llama-3.2-1B-Instruct-Q6_K.gguf",
186
  "size": 1021800576,
187
  "sha256": "0f4c510daf16e0d1b3bc94931fd9296c28936bebdda2593687d4eb70c5b70628",
188
  "required": false
189
  }
190
  ]
191
  },
192
- {
193
- "id": "gemma_4_uncensored",
194
- "name": "Gemma-4 Uncensored LLM",
195
- "category": "llm",
196
- "version": "1.0.0",
197
- "files": [
198
- {
199
- "id": "llm_gemma_4_e2b_uncensored_aggressive_q2_k_p",
200
- "path": "llm/gemma4/Gemma-4-E2B-Uncensored-HauhauCS-Aggressive-Q2_K_P.gguf",
201
- "size": 3009448896,
202
- "sha256": "c9213a33b16d513482790bb7092243f783a450e35f7deca584a41e0470afdb97",
203
- "required": false
204
- }
205
- ]
206
- },
207
  {
208
  "id": "supertonic_tts",
209
  "name": "Supertonic 3 TTS Engine",
@@ -269,14 +255,14 @@
269
  "files": [
270
  {
271
  "id": "tts_chatterbox_t3",
272
- "path": "tts/chatterbox/chatterbox-t3-mtl-q4_0.gguf",
273
  "size": 344985408,
274
  "sha256": "a7c8cbb267a836697a39757ada25a8ab095712b74961af2b86012dbd296a3a7b",
275
  "required": false
276
  },
277
  {
278
  "id": "tts_chatterbox_s3gen",
279
- "path": "tts/chatterbox/chatterbox-s3gen-mtl-f16.gguf",
280
  "size": 1056431360,
281
  "sha256": "b25fe64c5b5287952c82684f6cadb12698257dfe8ed3d9e578801ca970b59664",
282
  "required": false
@@ -293,7 +279,7 @@
293
  },
294
  {
295
  "id": "bge_m3",
296
- "name": "BGE-M3 Multilingual Embedding Engine",
297
  "category": "embedding",
298
  "version": "1.0.0",
299
  "files": [
@@ -302,13 +288,35 @@
302
  "path": "embedding/bge-m3/model_quantized.onnx",
303
  "size": 569694530,
304
  "sha256": "0826f8c1ab9edf1801db86c61919d4d108e8bfc0b809ec823ad366882ff0b77d",
305
- "required": true
306
  },
307
  {
308
  "id": "embedding_bge_m3_tokenizer",
309
  "path": "embedding/bge-m3/tokenizer.json",
310
  "size": 17082821,
311
  "sha256": "6710678b12670bc442b99edc952c4d996ae309a7020c1fa0096dd245c2faf790",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
312
  "required": true
313
  }
314
  ]
@@ -321,14 +329,14 @@
321
  "files": [
322
  {
323
  "id": "nli_deberta_onnx",
324
- "path": "nli/deberta-v3-xsmall-nli/model_quantized.onnx",
325
  "size": 87246587,
326
  "sha256": "3fac2500c45c75af42c7711de0d1b93d59577456100208be0dc1f9e8811946b6",
327
  "required": true
328
  },
329
  {
330
  "id": "nli_deberta_tokenizer",
331
- "path": "nli/deberta-v3-xsmall-nli/tokenizer.json",
332
  "size": 8656551,
333
  "sha256": "a86f883318afa11c8c10466f1bf4efaeb6ded28a52cbe57217a8fa0d0a2a87df",
334
  "required": true
 
1
  {
2
+ "models_version": "1.2.0",
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
+ "Nemotron 3.5, Qwen3 ASR, DeBERTa-v3 NLI, Supertonic 3, and Chatterbox TTS paths standardized."
8
  ],
9
  "total_size_bytes": 10081171563,
10
  "model_groups": [
 
67
  "files": [
68
  {
69
  "id": "stt_nemotron_config",
70
+ "path": "stt/nemotron-3.5/config.json",
71
  "size": 602,
72
  "sha256": "0592a224f30670e00f5e0f1f106f0171428b0abde0fd4cbc64237b02c2241502",
73
  "required": true
74
  },
75
  {
76
  "id": "stt_nemotron_decoder_joint",
77
+ "path": "stt/nemotron-3.5/decoder_joint.onnx",
78
  "size": 97590054,
79
  "sha256": "634dfadf24cb4f73c2fae170b36611d68db48186426882cbc8f7e02ed9f2bb29",
80
  "required": true
81
  },
82
  {
83
  "id": "stt_nemotron_encoder",
84
+ "path": "stt/nemotron-3.5/encoder.onnx",
85
  "size": 657572215,
86
  "sha256": "e653d494a4855ee87f6251e8a9677416a73edf6f36578841316fa2b77a53cda7",
87
  "required": true
88
  },
89
  {
90
  "id": "stt_nemotron_tokenizer",
91
+ "path": "stt/nemotron-3.5/tokenizer.model",
92
  "size": 406554,
93
  "sha256": "ce3895e40806f02a26c3a225161b96ef682d6c0054bae32a245dec4258d7d291",
94
  "required": true
 
153
  "files": [
154
  {
155
  "id": "llm_gemma_4_q4_k_m",
156
+ "path": "llm/gemma4/gemma-4-e2b-q4_k_m.gguf",
157
  "size": 3462678272,
158
  "sha256": "b5310340b3a23d31655d7119d100d5df1b2d8ee17b3ca8b0a23ad7e9eb5fa705",
159
  "required": false
 
168
  "files": [
169
  {
170
  "id": "llm_llama_3_2_1b_instruct_q4_k_m",
171
+ "path": "llm/llama/llama-3.2-1b-q4_k_m.gguf",
172
  "size": 807694464,
173
  "sha256": "6f85a640a97cf2bf5b8e764087b1e83da0fdb51d7c9fab7d0fece9385611df83",
174
  "required": false
 
183
  "files": [
184
  {
185
  "id": "llm_llama_3_2_1b_instruct_q6_k",
186
+ "path": "llm/llama/llama-3.2-1b-q6_k.gguf",
187
  "size": 1021800576,
188
  "sha256": "0f4c510daf16e0d1b3bc94931fd9296c28936bebdda2593687d4eb70c5b70628",
189
  "required": false
190
  }
191
  ]
192
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
193
  {
194
  "id": "supertonic_tts",
195
  "name": "Supertonic 3 TTS Engine",
 
255
  "files": [
256
  {
257
  "id": "tts_chatterbox_t3",
258
+ "path": "tts/chatterbox/t3-q4_0.gguf",
259
  "size": 344985408,
260
  "sha256": "a7c8cbb267a836697a39757ada25a8ab095712b74961af2b86012dbd296a3a7b",
261
  "required": false
262
  },
263
  {
264
  "id": "tts_chatterbox_s3gen",
265
+ "path": "tts/chatterbox/s3gen-f16.gguf",
266
  "size": 1056431360,
267
  "sha256": "b25fe64c5b5287952c82684f6cadb12698257dfe8ed3d9e578801ca970b59664",
268
  "required": false
 
279
  },
280
  {
281
  "id": "bge_m3",
282
+ "name": "BGE-M3 Multilingual Embedding Engine (Fallback)",
283
  "category": "embedding",
284
  "version": "1.0.0",
285
  "files": [
 
288
  "path": "embedding/bge-m3/model_quantized.onnx",
289
  "size": 569694530,
290
  "sha256": "0826f8c1ab9edf1801db86c61919d4d108e8bfc0b809ec823ad366882ff0b77d",
291
+ "required": false
292
  },
293
  {
294
  "id": "embedding_bge_m3_tokenizer",
295
  "path": "embedding/bge-m3/tokenizer.json",
296
  "size": 17082821,
297
  "sha256": "6710678b12670bc442b99edc952c4d996ae309a7020c1fa0096dd245c2faf790",
298
+ "required": false
299
+ }
300
+ ]
301
+ },
302
+ {
303
+ "id": "minilm_l12_v2",
304
+ "name": "MiniLM-L12 Multilingual Embedding Engine (Primary)",
305
+ "category": "embedding",
306
+ "version": "1.0.0",
307
+ "files": [
308
+ {
309
+ "id": "embedding_minilm_l12_onnx",
310
+ "path": "embedding/minilm-l12-v2/model_int8.onnx",
311
+ "size": 118054609,
312
+ "sha256": "4b76e8b4e72390a880356c38be28b9d363d6642d99dcf5c0c978bb7ecb3df637",
313
+ "required": true
314
+ },
315
+ {
316
+ "id": "embedding_minilm_l12_tokenizer",
317
+ "path": "embedding/minilm-l12-v2/tokenizer.json",
318
+ "size": 17082821,
319
+ "sha256": "6710678b12670bc442b99edc952c4d996ae309a7020c1fa0096dd245c2faf790",
320
  "required": true
321
  }
322
  ]
 
329
  "files": [
330
  {
331
  "id": "nli_deberta_onnx",
332
+ "path": "nli/deberta-v3-xsmall/model_quantized.onnx",
333
  "size": 87246587,
334
  "sha256": "3fac2500c45c75af42c7711de0d1b93d59577456100208be0dc1f9e8811946b6",
335
  "required": true
336
  },
337
  {
338
  "id": "nli_deberta_tokenizer",
339
+ "path": "nli/deberta-v3-xsmall/tokenizer.json",
340
  "size": 8656551,
341
  "sha256": "a86f883318afa11c8c10466f1bf4efaeb6ded28a52cbe57217a8fa0d0a2a87df",
342
  "required": true
nli/{deberta-v3-xsmall-nli → deberta-v3-xsmall}/model_quantized.onnx RENAMED
File without changes
nli/{deberta-v3-xsmall-nli → deberta-v3-xsmall}/tokenizer.json RENAMED
File without changes
stt/{nvidia-nemotron-3.5 → nemotron-3.5}/config.json RENAMED
File without changes
stt/{nvidia-nemotron-3.5 → nemotron-3.5}/decoder_joint.onnx RENAMED
File without changes
stt/{nvidia-nemotron-3.5 → nemotron-3.5}/encoder.onnx RENAMED
File without changes
stt/{nvidia-nemotron-3.5 → nemotron-3.5}/tokenizer.model RENAMED
File without changes
tts/chatterbox/{chatterbox-s3gen-mtl-f16.gguf → s3gen-f16.gguf} RENAMED
File without changes
tts/chatterbox/{chatterbox-t3-mtl-q4_0.gguf → t3-q4_0.gguf} RENAMED
File without changes