OzzyGT HF Staff commited on
Commit
9b2504b
·
verified ·
1 Parent(s): 6eff555

Upload 16 files

Browse files
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ pe_tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
model_index.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "ErnieImagePipeline",
3
+ "_diffusers_version": "0.36.0",
4
+ "scheduler": [
5
+ "diffusers",
6
+ "FlowMatchEulerDiscreteScheduler"
7
+ ],
8
+ "pe": [
9
+ "transformers",
10
+ "Ministral3ForCausalLM"
11
+ ],
12
+ "pe_tokenizer": [
13
+ "transformers",
14
+ "TokenizersBackend"
15
+ ],
16
+ "text_encoder": [
17
+ "transformers",
18
+ "Mistral3Model"
19
+ ],
20
+ "tokenizer": [
21
+ "transformers",
22
+ "TokenizersBackend"
23
+ ],
24
+ "transformer": [
25
+ "diffusers",
26
+ "ErnieImageTransformer2DModel"
27
+ ],
28
+ "vae": [
29
+ "diffusers",
30
+ "AutoencoderKLFlux2"
31
+ ]
32
+ }
pe/config.json ADDED
@@ -0,0 +1,281 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Ministral3ForCausalLM"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": 1,
7
+ "dtype": "bfloat16",
8
+ "eos_token_id": 2,
9
+ "head_dim": 128,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 3072,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 9216,
14
+ "max_position_embeddings": 262144,
15
+ "model_type": "ministral3",
16
+ "num_attention_heads": 32,
17
+ "num_hidden_layers": 26,
18
+ "num_key_value_heads": 8,
19
+ "pad_token_id": 11,
20
+ "quantization_config": {
21
+ "add_skip_keys": true,
22
+ "dequantize_fp32": false,
23
+ "dynamic_loss_threshold": 0.0001,
24
+ "group_size": 0,
25
+ "is_integer": true,
26
+ "is_training": false,
27
+ "modules_dtype_dict": {
28
+ "int8": [
29
+ "model.layers.0.mlp.down_proj.weight",
30
+ "model.layers.0.mlp.gate_proj.weight",
31
+ "model.layers.0.mlp.up_proj.weight",
32
+ "model.layers.0.self_attn.k_proj.weight",
33
+ "model.layers.0.self_attn.q_proj.weight",
34
+ "model.layers.0.self_attn.v_proj.weight",
35
+ "model.layers.1.mlp.down_proj.weight",
36
+ "model.layers.1.mlp.gate_proj.weight",
37
+ "model.layers.1.mlp.up_proj.weight",
38
+ "model.layers.1.self_attn.k_proj.weight",
39
+ "model.layers.1.self_attn.o_proj.weight",
40
+ "model.layers.1.self_attn.q_proj.weight",
41
+ "model.layers.1.self_attn.v_proj.weight",
42
+ "model.layers.2.mlp.down_proj.weight",
43
+ "model.layers.2.mlp.gate_proj.weight",
44
+ "model.layers.2.mlp.up_proj.weight",
45
+ "model.layers.2.self_attn.k_proj.weight",
46
+ "model.layers.2.self_attn.o_proj.weight",
47
+ "model.layers.2.self_attn.q_proj.weight",
48
+ "model.layers.2.self_attn.v_proj.weight",
49
+ "model.layers.3.mlp.down_proj.weight",
50
+ "model.layers.3.mlp.gate_proj.weight",
51
+ "model.layers.3.mlp.up_proj.weight",
52
+ "model.layers.3.self_attn.k_proj.weight",
53
+ "model.layers.3.self_attn.o_proj.weight",
54
+ "model.layers.3.self_attn.q_proj.weight",
55
+ "model.layers.3.self_attn.v_proj.weight",
56
+ "model.layers.4.mlp.down_proj.weight",
57
+ "model.layers.4.mlp.gate_proj.weight",
58
+ "model.layers.4.mlp.up_proj.weight",
59
+ "model.layers.4.self_attn.k_proj.weight",
60
+ "model.layers.4.self_attn.o_proj.weight",
61
+ "model.layers.4.self_attn.q_proj.weight",
62
+ "model.layers.4.self_attn.v_proj.weight",
63
+ "model.layers.5.mlp.down_proj.weight",
64
+ "model.layers.5.mlp.gate_proj.weight",
65
+ "model.layers.5.mlp.up_proj.weight",
66
+ "model.layers.5.self_attn.k_proj.weight",
67
+ "model.layers.5.self_attn.o_proj.weight",
68
+ "model.layers.5.self_attn.q_proj.weight",
69
+ "model.layers.5.self_attn.v_proj.weight",
70
+ "model.layers.6.mlp.down_proj.weight",
71
+ "model.layers.6.mlp.gate_proj.weight",
72
+ "model.layers.6.mlp.up_proj.weight",
73
+ "model.layers.6.self_attn.k_proj.weight",
74
+ "model.layers.6.self_attn.o_proj.weight",
75
+ "model.layers.6.self_attn.q_proj.weight",
76
+ "model.layers.6.self_attn.v_proj.weight",
77
+ "model.layers.7.mlp.down_proj.weight",
78
+ "model.layers.7.mlp.gate_proj.weight",
79
+ "model.layers.7.mlp.up_proj.weight",
80
+ "model.layers.7.self_attn.k_proj.weight",
81
+ "model.layers.7.self_attn.o_proj.weight",
82
+ "model.layers.7.self_attn.q_proj.weight",
83
+ "model.layers.7.self_attn.v_proj.weight",
84
+ "model.layers.8.mlp.down_proj.weight",
85
+ "model.layers.8.mlp.gate_proj.weight",
86
+ "model.layers.8.mlp.up_proj.weight",
87
+ "model.layers.8.self_attn.k_proj.weight",
88
+ "model.layers.8.self_attn.o_proj.weight",
89
+ "model.layers.8.self_attn.q_proj.weight",
90
+ "model.layers.8.self_attn.v_proj.weight",
91
+ "model.layers.9.mlp.down_proj.weight",
92
+ "model.layers.9.mlp.gate_proj.weight",
93
+ "model.layers.9.mlp.up_proj.weight",
94
+ "model.layers.9.self_attn.k_proj.weight",
95
+ "model.layers.9.self_attn.o_proj.weight",
96
+ "model.layers.9.self_attn.q_proj.weight",
97
+ "model.layers.9.self_attn.v_proj.weight",
98
+ "model.layers.10.mlp.down_proj.weight",
99
+ "model.layers.10.mlp.gate_proj.weight",
100
+ "model.layers.10.mlp.up_proj.weight",
101
+ "model.layers.10.self_attn.k_proj.weight",
102
+ "model.layers.10.self_attn.o_proj.weight",
103
+ "model.layers.10.self_attn.q_proj.weight",
104
+ "model.layers.10.self_attn.v_proj.weight",
105
+ "model.layers.11.mlp.down_proj.weight",
106
+ "model.layers.11.mlp.gate_proj.weight",
107
+ "model.layers.11.mlp.up_proj.weight",
108
+ "model.layers.11.self_attn.k_proj.weight",
109
+ "model.layers.11.self_attn.o_proj.weight",
110
+ "model.layers.11.self_attn.q_proj.weight",
111
+ "model.layers.11.self_attn.v_proj.weight",
112
+ "model.layers.12.mlp.down_proj.weight",
113
+ "model.layers.12.mlp.gate_proj.weight",
114
+ "model.layers.12.mlp.up_proj.weight",
115
+ "model.layers.12.self_attn.k_proj.weight",
116
+ "model.layers.12.self_attn.o_proj.weight",
117
+ "model.layers.12.self_attn.q_proj.weight",
118
+ "model.layers.12.self_attn.v_proj.weight",
119
+ "model.layers.13.mlp.down_proj.weight",
120
+ "model.layers.13.mlp.gate_proj.weight",
121
+ "model.layers.13.mlp.up_proj.weight",
122
+ "model.layers.13.self_attn.k_proj.weight",
123
+ "model.layers.13.self_attn.o_proj.weight",
124
+ "model.layers.13.self_attn.q_proj.weight",
125
+ "model.layers.13.self_attn.v_proj.weight",
126
+ "model.layers.14.mlp.down_proj.weight",
127
+ "model.layers.14.mlp.gate_proj.weight",
128
+ "model.layers.14.mlp.up_proj.weight",
129
+ "model.layers.14.self_attn.k_proj.weight",
130
+ "model.layers.14.self_attn.o_proj.weight",
131
+ "model.layers.14.self_attn.q_proj.weight",
132
+ "model.layers.14.self_attn.v_proj.weight",
133
+ "model.layers.15.mlp.down_proj.weight",
134
+ "model.layers.15.mlp.gate_proj.weight",
135
+ "model.layers.15.mlp.up_proj.weight",
136
+ "model.layers.15.self_attn.k_proj.weight",
137
+ "model.layers.15.self_attn.o_proj.weight",
138
+ "model.layers.15.self_attn.q_proj.weight",
139
+ "model.layers.15.self_attn.v_proj.weight",
140
+ "model.layers.16.mlp.down_proj.weight",
141
+ "model.layers.16.mlp.gate_proj.weight",
142
+ "model.layers.16.mlp.up_proj.weight",
143
+ "model.layers.16.self_attn.k_proj.weight",
144
+ "model.layers.16.self_attn.o_proj.weight",
145
+ "model.layers.16.self_attn.q_proj.weight",
146
+ "model.layers.16.self_attn.v_proj.weight",
147
+ "model.layers.17.mlp.down_proj.weight",
148
+ "model.layers.17.mlp.gate_proj.weight",
149
+ "model.layers.17.mlp.up_proj.weight",
150
+ "model.layers.17.self_attn.k_proj.weight",
151
+ "model.layers.17.self_attn.o_proj.weight",
152
+ "model.layers.17.self_attn.q_proj.weight",
153
+ "model.layers.17.self_attn.v_proj.weight",
154
+ "model.layers.18.mlp.down_proj.weight",
155
+ "model.layers.18.mlp.gate_proj.weight",
156
+ "model.layers.18.mlp.up_proj.weight",
157
+ "model.layers.18.self_attn.k_proj.weight",
158
+ "model.layers.18.self_attn.o_proj.weight",
159
+ "model.layers.18.self_attn.q_proj.weight",
160
+ "model.layers.18.self_attn.v_proj.weight",
161
+ "model.layers.19.mlp.down_proj.weight",
162
+ "model.layers.19.mlp.gate_proj.weight",
163
+ "model.layers.19.mlp.up_proj.weight",
164
+ "model.layers.19.self_attn.k_proj.weight",
165
+ "model.layers.19.self_attn.o_proj.weight",
166
+ "model.layers.19.self_attn.q_proj.weight",
167
+ "model.layers.19.self_attn.v_proj.weight",
168
+ "model.layers.20.mlp.down_proj.weight",
169
+ "model.layers.20.mlp.gate_proj.weight",
170
+ "model.layers.20.mlp.up_proj.weight",
171
+ "model.layers.20.self_attn.k_proj.weight",
172
+ "model.layers.20.self_attn.o_proj.weight",
173
+ "model.layers.20.self_attn.q_proj.weight",
174
+ "model.layers.20.self_attn.v_proj.weight",
175
+ "model.layers.21.mlp.down_proj.weight",
176
+ "model.layers.21.mlp.gate_proj.weight",
177
+ "model.layers.21.mlp.up_proj.weight",
178
+ "model.layers.21.self_attn.k_proj.weight",
179
+ "model.layers.21.self_attn.o_proj.weight",
180
+ "model.layers.21.self_attn.q_proj.weight",
181
+ "model.layers.21.self_attn.v_proj.weight",
182
+ "model.layers.22.mlp.down_proj.weight",
183
+ "model.layers.22.mlp.gate_proj.weight",
184
+ "model.layers.22.mlp.up_proj.weight",
185
+ "model.layers.22.self_attn.k_proj.weight",
186
+ "model.layers.22.self_attn.o_proj.weight",
187
+ "model.layers.22.self_attn.q_proj.weight",
188
+ "model.layers.22.self_attn.v_proj.weight",
189
+ "model.layers.23.mlp.down_proj.weight",
190
+ "model.layers.23.mlp.gate_proj.weight",
191
+ "model.layers.23.mlp.up_proj.weight",
192
+ "model.layers.23.self_attn.k_proj.weight",
193
+ "model.layers.23.self_attn.o_proj.weight",
194
+ "model.layers.23.self_attn.q_proj.weight",
195
+ "model.layers.23.self_attn.v_proj.weight",
196
+ "model.layers.24.mlp.down_proj.weight",
197
+ "model.layers.24.mlp.gate_proj.weight",
198
+ "model.layers.24.mlp.up_proj.weight",
199
+ "model.layers.24.self_attn.k_proj.weight",
200
+ "model.layers.24.self_attn.o_proj.weight",
201
+ "model.layers.24.self_attn.q_proj.weight",
202
+ "model.layers.24.self_attn.v_proj.weight",
203
+ "model.layers.25.mlp.down_proj.weight",
204
+ "model.layers.25.mlp.gate_proj.weight",
205
+ "model.layers.25.mlp.up_proj.weight",
206
+ "model.layers.25.self_attn.k_proj.weight",
207
+ "model.layers.25.self_attn.o_proj.weight",
208
+ "model.layers.25.self_attn.q_proj.weight",
209
+ "model.layers.25.self_attn.v_proj.weight"
210
+ ],
211
+ "uint8": [
212
+ "model.layers.0.self_attn.o_proj.weight"
213
+ ]
214
+ },
215
+ "modules_quant_config": {},
216
+ "modules_to_not_convert": [
217
+ "lm_head.weight",
218
+ "lm_head",
219
+ "time_text_embed",
220
+ ".proj_out",
221
+ ".emb_out",
222
+ ".norm_out",
223
+ "model.embed_tokens",
224
+ ".vid_in",
225
+ ".txt_in",
226
+ ".x_embedder",
227
+ ".time_embed",
228
+ ".img_out",
229
+ ".txt_out",
230
+ "model.embed_tokens.weight",
231
+ "patch_embed",
232
+ "patch_embedding",
233
+ ".emb_in",
234
+ ".t_embedder",
235
+ ".condition_embedder",
236
+ ".context_embedder",
237
+ ".vid_out",
238
+ "multi_modal_projector",
239
+ ".y_embedder",
240
+ "patch_emb",
241
+ "wte",
242
+ ".img_in",
243
+ ".final_layer"
244
+ ],
245
+ "non_blocking": false,
246
+ "quant_conv": false,
247
+ "quant_method": "sdnq",
248
+ "quantization_device": null,
249
+ "quantized_matmul_dtype": null,
250
+ "return_device": null,
251
+ "sdnq_version": "0.1.6",
252
+ "svd_rank": 32,
253
+ "svd_steps": 8,
254
+ "use_dynamic_quantization": true,
255
+ "use_grad_ckpt": true,
256
+ "use_quantized_matmul": false,
257
+ "use_quantized_matmul_conv": false,
258
+ "use_static_quantization": true,
259
+ "use_stochastic_rounding": false,
260
+ "use_svd": false,
261
+ "weights_dtype": "int8"
262
+ },
263
+ "rms_norm_eps": 1e-05,
264
+ "rope_parameters": {
265
+ "beta_fast": 32.0,
266
+ "beta_slow": 1.0,
267
+ "factor": 16.0,
268
+ "llama_4_scaling_beta": 0.1,
269
+ "mscale": 1.0,
270
+ "mscale_all_dim": 1.0,
271
+ "original_max_position_embeddings": 16384,
272
+ "rope_theta": 1000000.0,
273
+ "rope_type": "yarn",
274
+ "type": "yarn"
275
+ },
276
+ "sliding_window": null,
277
+ "tie_word_embeddings": false,
278
+ "transformers_version": "5.3.0",
279
+ "use_cache": true,
280
+ "vocab_size": 131072
281
+ }
pe/generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 11,
6
+ "transformers_version": "5.3.0",
7
+ "use_cache": true
8
+ }
pe/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1807b3af336a21c7614fb3d88c3567f6e0196300f2e7fb857f443b98e9c89d64
3
+ size 4643112168
pe_tokenizer/chat_template.jinja ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {{- bos_token }}[SYSTEM_PROMPT]你是一个专业的文生图 Prompt 增强助手。你将收到用户的简短图片描述及目标生成分辨率,请据此扩写为一段内容丰富、细节充分的视觉描述,以帮助文生图模型生成高质量的图片。仅输出增强后的描述,不要包含任何解释或前缀。[/SYSTEM_PROMPT]
2
+ {%- for message in messages %}
3
+ {%- if message['role'] == 'user' %}
4
+ {{- '[INST]' + message['content'] + '[/INST]' }}
5
+ {%- elif message['role'] == 'assistant' %}
6
+ {%- generation %}{{ message['content'] }}{{ eos_token }}{% endgeneration %}
7
+ {%- endif %}
8
+ {%- endfor %}
pe_tokenizer/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:286acad9b0e27fce778ac429763536accf618ccb6ed72963b6f94685e531c5c7
3
+ size 17077402
pe_tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,1013 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "extra_special_tokens": [
6
+ "<unk>",
7
+ "<s>",
8
+ "</s>",
9
+ "[INST]",
10
+ "[/INST]",
11
+ "[AVAILABLE_TOOLS]",
12
+ "[/AVAILABLE_TOOLS]",
13
+ "[TOOL_RESULTS]",
14
+ "[/TOOL_RESULTS]",
15
+ "[TOOL_CALLS]",
16
+ "[IMG]",
17
+ "<pad>",
18
+ "[IMG_BREAK]",
19
+ "[IMG_END]",
20
+ "[PREFIX]",
21
+ "[MIDDLE]",
22
+ "[SUFFIX]",
23
+ "[SYSTEM_PROMPT]",
24
+ "[/SYSTEM_PROMPT]",
25
+ "[TOOL_CONTENT]",
26
+ "<SPECIAL_20>",
27
+ "<SPECIAL_21>",
28
+ "<SPECIAL_22>",
29
+ "<SPECIAL_23>",
30
+ "[AUDIO]",
31
+ "[BEGIN_AUDIO]",
32
+ "<SPECIAL_26>",
33
+ "<SPECIAL_27>",
34
+ "<SPECIAL_28>",
35
+ "<SPECIAL_29>",
36
+ "<SPECIAL_30>",
37
+ "<SPECIAL_31>",
38
+ "[ARGS]",
39
+ "[CALL_ID]",
40
+ "[THINK]",
41
+ "[/THINK]",
42
+ "<SPECIAL_36>",
43
+ "<SPECIAL_37>",
44
+ "<SPECIAL_38>",
45
+ "<SPECIAL_39>",
46
+ "<SPECIAL_40>",
47
+ "<SPECIAL_41>",
48
+ "<SPECIAL_42>",
49
+ "<SPECIAL_43>",
50
+ "<SPECIAL_44>",
51
+ "<SPECIAL_45>",
52
+ "<SPECIAL_46>",
53
+ "<SPECIAL_47>",
54
+ "<SPECIAL_48>",
55
+ "<SPECIAL_49>",
56
+ "<SPECIAL_50>",
57
+ "<SPECIAL_51>",
58
+ "<SPECIAL_52>",
59
+ "<SPECIAL_53>",
60
+ "<SPECIAL_54>",
61
+ "<SPECIAL_55>",
62
+ "<SPECIAL_56>",
63
+ "<SPECIAL_57>",
64
+ "<SPECIAL_58>",
65
+ "<SPECIAL_59>",
66
+ "<SPECIAL_60>",
67
+ "<SPECIAL_61>",
68
+ "<SPECIAL_62>",
69
+ "<SPECIAL_63>",
70
+ "<SPECIAL_64>",
71
+ "<SPECIAL_65>",
72
+ "<SPECIAL_66>",
73
+ "<SPECIAL_67>",
74
+ "<SPECIAL_68>",
75
+ "<SPECIAL_69>",
76
+ "<SPECIAL_70>",
77
+ "<SPECIAL_71>",
78
+ "<SPECIAL_72>",
79
+ "<SPECIAL_73>",
80
+ "<SPECIAL_74>",
81
+ "<SPECIAL_75>",
82
+ "<SPECIAL_76>",
83
+ "<SPECIAL_77>",
84
+ "<SPECIAL_78>",
85
+ "<SPECIAL_79>",
86
+ "<SPECIAL_80>",
87
+ "<SPECIAL_81>",
88
+ "<SPECIAL_82>",
89
+ "<SPECIAL_83>",
90
+ "<SPECIAL_84>",
91
+ "<SPECIAL_85>",
92
+ "<SPECIAL_86>",
93
+ "<SPECIAL_87>",
94
+ "<SPECIAL_88>",
95
+ "<SPECIAL_89>",
96
+ "<SPECIAL_90>",
97
+ "<SPECIAL_91>",
98
+ "<SPECIAL_92>",
99
+ "<SPECIAL_93>",
100
+ "<SPECIAL_94>",
101
+ "<SPECIAL_95>",
102
+ "<SPECIAL_96>",
103
+ "<SPECIAL_97>",
104
+ "<SPECIAL_98>",
105
+ "<SPECIAL_99>",
106
+ "<SPECIAL_100>",
107
+ "<SPECIAL_101>",
108
+ "<SPECIAL_102>",
109
+ "<SPECIAL_103>",
110
+ "<SPECIAL_104>",
111
+ "<SPECIAL_105>",
112
+ "<SPECIAL_106>",
113
+ "<SPECIAL_107>",
114
+ "<SPECIAL_108>",
115
+ "<SPECIAL_109>",
116
+ "<SPECIAL_110>",
117
+ "<SPECIAL_111>",
118
+ "<SPECIAL_112>",
119
+ "<SPECIAL_113>",
120
+ "<SPECIAL_114>",
121
+ "<SPECIAL_115>",
122
+ "<SPECIAL_116>",
123
+ "<SPECIAL_117>",
124
+ "<SPECIAL_118>",
125
+ "<SPECIAL_119>",
126
+ "<SPECIAL_120>",
127
+ "<SPECIAL_121>",
128
+ "<SPECIAL_122>",
129
+ "<SPECIAL_123>",
130
+ "<SPECIAL_124>",
131
+ "<SPECIAL_125>",
132
+ "<SPECIAL_126>",
133
+ "<SPECIAL_127>",
134
+ "<SPECIAL_128>",
135
+ "<SPECIAL_129>",
136
+ "<SPECIAL_130>",
137
+ "<SPECIAL_131>",
138
+ "<SPECIAL_132>",
139
+ "<SPECIAL_133>",
140
+ "<SPECIAL_134>",
141
+ "<SPECIAL_135>",
142
+ "<SPECIAL_136>",
143
+ "<SPECIAL_137>",
144
+ "<SPECIAL_138>",
145
+ "<SPECIAL_139>",
146
+ "<SPECIAL_140>",
147
+ "<SPECIAL_141>",
148
+ "<SPECIAL_142>",
149
+ "<SPECIAL_143>",
150
+ "<SPECIAL_144>",
151
+ "<SPECIAL_145>",
152
+ "<SPECIAL_146>",
153
+ "<SPECIAL_147>",
154
+ "<SPECIAL_148>",
155
+ "<SPECIAL_149>",
156
+ "<SPECIAL_150>",
157
+ "<SPECIAL_151>",
158
+ "<SPECIAL_152>",
159
+ "<SPECIAL_153>",
160
+ "<SPECIAL_154>",
161
+ "<SPECIAL_155>",
162
+ "<SPECIAL_156>",
163
+ "<SPECIAL_157>",
164
+ "<SPECIAL_158>",
165
+ "<SPECIAL_159>",
166
+ "<SPECIAL_160>",
167
+ "<SPECIAL_161>",
168
+ "<SPECIAL_162>",
169
+ "<SPECIAL_163>",
170
+ "<SPECIAL_164>",
171
+ "<SPECIAL_165>",
172
+ "<SPECIAL_166>",
173
+ "<SPECIAL_167>",
174
+ "<SPECIAL_168>",
175
+ "<SPECIAL_169>",
176
+ "<SPECIAL_170>",
177
+ "<SPECIAL_171>",
178
+ "<SPECIAL_172>",
179
+ "<SPECIAL_173>",
180
+ "<SPECIAL_174>",
181
+ "<SPECIAL_175>",
182
+ "<SPECIAL_176>",
183
+ "<SPECIAL_177>",
184
+ "<SPECIAL_178>",
185
+ "<SPECIAL_179>",
186
+ "<SPECIAL_180>",
187
+ "<SPECIAL_181>",
188
+ "<SPECIAL_182>",
189
+ "<SPECIAL_183>",
190
+ "<SPECIAL_184>",
191
+ "<SPECIAL_185>",
192
+ "<SPECIAL_186>",
193
+ "<SPECIAL_187>",
194
+ "<SPECIAL_188>",
195
+ "<SPECIAL_189>",
196
+ "<SPECIAL_190>",
197
+ "<SPECIAL_191>",
198
+ "<SPECIAL_192>",
199
+ "<SPECIAL_193>",
200
+ "<SPECIAL_194>",
201
+ "<SPECIAL_195>",
202
+ "<SPECIAL_196>",
203
+ "<SPECIAL_197>",
204
+ "<SPECIAL_198>",
205
+ "<SPECIAL_199>",
206
+ "<SPECIAL_200>",
207
+ "<SPECIAL_201>",
208
+ "<SPECIAL_202>",
209
+ "<SPECIAL_203>",
210
+ "<SPECIAL_204>",
211
+ "<SPECIAL_205>",
212
+ "<SPECIAL_206>",
213
+ "<SPECIAL_207>",
214
+ "<SPECIAL_208>",
215
+ "<SPECIAL_209>",
216
+ "<SPECIAL_210>",
217
+ "<SPECIAL_211>",
218
+ "<SPECIAL_212>",
219
+ "<SPECIAL_213>",
220
+ "<SPECIAL_214>",
221
+ "<SPECIAL_215>",
222
+ "<SPECIAL_216>",
223
+ "<SPECIAL_217>",
224
+ "<SPECIAL_218>",
225
+ "<SPECIAL_219>",
226
+ "<SPECIAL_220>",
227
+ "<SPECIAL_221>",
228
+ "<SPECIAL_222>",
229
+ "<SPECIAL_223>",
230
+ "<SPECIAL_224>",
231
+ "<SPECIAL_225>",
232
+ "<SPECIAL_226>",
233
+ "<SPECIAL_227>",
234
+ "<SPECIAL_228>",
235
+ "<SPECIAL_229>",
236
+ "<SPECIAL_230>",
237
+ "<SPECIAL_231>",
238
+ "<SPECIAL_232>",
239
+ "<SPECIAL_233>",
240
+ "<SPECIAL_234>",
241
+ "<SPECIAL_235>",
242
+ "<SPECIAL_236>",
243
+ "<SPECIAL_237>",
244
+ "<SPECIAL_238>",
245
+ "<SPECIAL_239>",
246
+ "<SPECIAL_240>",
247
+ "<SPECIAL_241>",
248
+ "<SPECIAL_242>",
249
+ "<SPECIAL_243>",
250
+ "<SPECIAL_244>",
251
+ "<SPECIAL_245>",
252
+ "<SPECIAL_246>",
253
+ "<SPECIAL_247>",
254
+ "<SPECIAL_248>",
255
+ "<SPECIAL_249>",
256
+ "<SPECIAL_250>",
257
+ "<SPECIAL_251>",
258
+ "<SPECIAL_252>",
259
+ "<SPECIAL_253>",
260
+ "<SPECIAL_254>",
261
+ "<SPECIAL_255>",
262
+ "<SPECIAL_256>",
263
+ "<SPECIAL_257>",
264
+ "<SPECIAL_258>",
265
+ "<SPECIAL_259>",
266
+ "<SPECIAL_260>",
267
+ "<SPECIAL_261>",
268
+ "<SPECIAL_262>",
269
+ "<SPECIAL_263>",
270
+ "<SPECIAL_264>",
271
+ "<SPECIAL_265>",
272
+ "<SPECIAL_266>",
273
+ "<SPECIAL_267>",
274
+ "<SPECIAL_268>",
275
+ "<SPECIAL_269>",
276
+ "<SPECIAL_270>",
277
+ "<SPECIAL_271>",
278
+ "<SPECIAL_272>",
279
+ "<SPECIAL_273>",
280
+ "<SPECIAL_274>",
281
+ "<SPECIAL_275>",
282
+ "<SPECIAL_276>",
283
+ "<SPECIAL_277>",
284
+ "<SPECIAL_278>",
285
+ "<SPECIAL_279>",
286
+ "<SPECIAL_280>",
287
+ "<SPECIAL_281>",
288
+ "<SPECIAL_282>",
289
+ "<SPECIAL_283>",
290
+ "<SPECIAL_284>",
291
+ "<SPECIAL_285>",
292
+ "<SPECIAL_286>",
293
+ "<SPECIAL_287>",
294
+ "<SPECIAL_288>",
295
+ "<SPECIAL_289>",
296
+ "<SPECIAL_290>",
297
+ "<SPECIAL_291>",
298
+ "<SPECIAL_292>",
299
+ "<SPECIAL_293>",
300
+ "<SPECIAL_294>",
301
+ "<SPECIAL_295>",
302
+ "<SPECIAL_296>",
303
+ "<SPECIAL_297>",
304
+ "<SPECIAL_298>",
305
+ "<SPECIAL_299>",
306
+ "<SPECIAL_300>",
307
+ "<SPECIAL_301>",
308
+ "<SPECIAL_302>",
309
+ "<SPECIAL_303>",
310
+ "<SPECIAL_304>",
311
+ "<SPECIAL_305>",
312
+ "<SPECIAL_306>",
313
+ "<SPECIAL_307>",
314
+ "<SPECIAL_308>",
315
+ "<SPECIAL_309>",
316
+ "<SPECIAL_310>",
317
+ "<SPECIAL_311>",
318
+ "<SPECIAL_312>",
319
+ "<SPECIAL_313>",
320
+ "<SPECIAL_314>",
321
+ "<SPECIAL_315>",
322
+ "<SPECIAL_316>",
323
+ "<SPECIAL_317>",
324
+ "<SPECIAL_318>",
325
+ "<SPECIAL_319>",
326
+ "<SPECIAL_320>",
327
+ "<SPECIAL_321>",
328
+ "<SPECIAL_322>",
329
+ "<SPECIAL_323>",
330
+ "<SPECIAL_324>",
331
+ "<SPECIAL_325>",
332
+ "<SPECIAL_326>",
333
+ "<SPECIAL_327>",
334
+ "<SPECIAL_328>",
335
+ "<SPECIAL_329>",
336
+ "<SPECIAL_330>",
337
+ "<SPECIAL_331>",
338
+ "<SPECIAL_332>",
339
+ "<SPECIAL_333>",
340
+ "<SPECIAL_334>",
341
+ "<SPECIAL_335>",
342
+ "<SPECIAL_336>",
343
+ "<SPECIAL_337>",
344
+ "<SPECIAL_338>",
345
+ "<SPECIAL_339>",
346
+ "<SPECIAL_340>",
347
+ "<SPECIAL_341>",
348
+ "<SPECIAL_342>",
349
+ "<SPECIAL_343>",
350
+ "<SPECIAL_344>",
351
+ "<SPECIAL_345>",
352
+ "<SPECIAL_346>",
353
+ "<SPECIAL_347>",
354
+ "<SPECIAL_348>",
355
+ "<SPECIAL_349>",
356
+ "<SPECIAL_350>",
357
+ "<SPECIAL_351>",
358
+ "<SPECIAL_352>",
359
+ "<SPECIAL_353>",
360
+ "<SPECIAL_354>",
361
+ "<SPECIAL_355>",
362
+ "<SPECIAL_356>",
363
+ "<SPECIAL_357>",
364
+ "<SPECIAL_358>",
365
+ "<SPECIAL_359>",
366
+ "<SPECIAL_360>",
367
+ "<SPECIAL_361>",
368
+ "<SPECIAL_362>",
369
+ "<SPECIAL_363>",
370
+ "<SPECIAL_364>",
371
+ "<SPECIAL_365>",
372
+ "<SPECIAL_366>",
373
+ "<SPECIAL_367>",
374
+ "<SPECIAL_368>",
375
+ "<SPECIAL_369>",
376
+ "<SPECIAL_370>",
377
+ "<SPECIAL_371>",
378
+ "<SPECIAL_372>",
379
+ "<SPECIAL_373>",
380
+ "<SPECIAL_374>",
381
+ "<SPECIAL_375>",
382
+ "<SPECIAL_376>",
383
+ "<SPECIAL_377>",
384
+ "<SPECIAL_378>",
385
+ "<SPECIAL_379>",
386
+ "<SPECIAL_380>",
387
+ "<SPECIAL_381>",
388
+ "<SPECIAL_382>",
389
+ "<SPECIAL_383>",
390
+ "<SPECIAL_384>",
391
+ "<SPECIAL_385>",
392
+ "<SPECIAL_386>",
393
+ "<SPECIAL_387>",
394
+ "<SPECIAL_388>",
395
+ "<SPECIAL_389>",
396
+ "<SPECIAL_390>",
397
+ "<SPECIAL_391>",
398
+ "<SPECIAL_392>",
399
+ "<SPECIAL_393>",
400
+ "<SPECIAL_394>",
401
+ "<SPECIAL_395>",
402
+ "<SPECIAL_396>",
403
+ "<SPECIAL_397>",
404
+ "<SPECIAL_398>",
405
+ "<SPECIAL_399>",
406
+ "<SPECIAL_400>",
407
+ "<SPECIAL_401>",
408
+ "<SPECIAL_402>",
409
+ "<SPECIAL_403>",
410
+ "<SPECIAL_404>",
411
+ "<SPECIAL_405>",
412
+ "<SPECIAL_406>",
413
+ "<SPECIAL_407>",
414
+ "<SPECIAL_408>",
415
+ "<SPECIAL_409>",
416
+ "<SPECIAL_410>",
417
+ "<SPECIAL_411>",
418
+ "<SPECIAL_412>",
419
+ "<SPECIAL_413>",
420
+ "<SPECIAL_414>",
421
+ "<SPECIAL_415>",
422
+ "<SPECIAL_416>",
423
+ "<SPECIAL_417>",
424
+ "<SPECIAL_418>",
425
+ "<SPECIAL_419>",
426
+ "<SPECIAL_420>",
427
+ "<SPECIAL_421>",
428
+ "<SPECIAL_422>",
429
+ "<SPECIAL_423>",
430
+ "<SPECIAL_424>",
431
+ "<SPECIAL_425>",
432
+ "<SPECIAL_426>",
433
+ "<SPECIAL_427>",
434
+ "<SPECIAL_428>",
435
+ "<SPECIAL_429>",
436
+ "<SPECIAL_430>",
437
+ "<SPECIAL_431>",
438
+ "<SPECIAL_432>",
439
+ "<SPECIAL_433>",
440
+ "<SPECIAL_434>",
441
+ "<SPECIAL_435>",
442
+ "<SPECIAL_436>",
443
+ "<SPECIAL_437>",
444
+ "<SPECIAL_438>",
445
+ "<SPECIAL_439>",
446
+ "<SPECIAL_440>",
447
+ "<SPECIAL_441>",
448
+ "<SPECIAL_442>",
449
+ "<SPECIAL_443>",
450
+ "<SPECIAL_444>",
451
+ "<SPECIAL_445>",
452
+ "<SPECIAL_446>",
453
+ "<SPECIAL_447>",
454
+ "<SPECIAL_448>",
455
+ "<SPECIAL_449>",
456
+ "<SPECIAL_450>",
457
+ "<SPECIAL_451>",
458
+ "<SPECIAL_452>",
459
+ "<SPECIAL_453>",
460
+ "<SPECIAL_454>",
461
+ "<SPECIAL_455>",
462
+ "<SPECIAL_456>",
463
+ "<SPECIAL_457>",
464
+ "<SPECIAL_458>",
465
+ "<SPECIAL_459>",
466
+ "<SPECIAL_460>",
467
+ "<SPECIAL_461>",
468
+ "<SPECIAL_462>",
469
+ "<SPECIAL_463>",
470
+ "<SPECIAL_464>",
471
+ "<SPECIAL_465>",
472
+ "<SPECIAL_466>",
473
+ "<SPECIAL_467>",
474
+ "<SPECIAL_468>",
475
+ "<SPECIAL_469>",
476
+ "<SPECIAL_470>",
477
+ "<SPECIAL_471>",
478
+ "<SPECIAL_472>",
479
+ "<SPECIAL_473>",
480
+ "<SPECIAL_474>",
481
+ "<SPECIAL_475>",
482
+ "<SPECIAL_476>",
483
+ "<SPECIAL_477>",
484
+ "<SPECIAL_478>",
485
+ "<SPECIAL_479>",
486
+ "<SPECIAL_480>",
487
+ "<SPECIAL_481>",
488
+ "<SPECIAL_482>",
489
+ "<SPECIAL_483>",
490
+ "<SPECIAL_484>",
491
+ "<SPECIAL_485>",
492
+ "<SPECIAL_486>",
493
+ "<SPECIAL_487>",
494
+ "<SPECIAL_488>",
495
+ "<SPECIAL_489>",
496
+ "<SPECIAL_490>",
497
+ "<SPECIAL_491>",
498
+ "<SPECIAL_492>",
499
+ "<SPECIAL_493>",
500
+ "<SPECIAL_494>",
501
+ "<SPECIAL_495>",
502
+ "<SPECIAL_496>",
503
+ "<SPECIAL_497>",
504
+ "<SPECIAL_498>",
505
+ "<SPECIAL_499>",
506
+ "<SPECIAL_500>",
507
+ "<SPECIAL_501>",
508
+ "<SPECIAL_502>",
509
+ "<SPECIAL_503>",
510
+ "<SPECIAL_504>",
511
+ "<SPECIAL_505>",
512
+ "<SPECIAL_506>",
513
+ "<SPECIAL_507>",
514
+ "<SPECIAL_508>",
515
+ "<SPECIAL_509>",
516
+ "<SPECIAL_510>",
517
+ "<SPECIAL_511>",
518
+ "<SPECIAL_512>",
519
+ "<SPECIAL_513>",
520
+ "<SPECIAL_514>",
521
+ "<SPECIAL_515>",
522
+ "<SPECIAL_516>",
523
+ "<SPECIAL_517>",
524
+ "<SPECIAL_518>",
525
+ "<SPECIAL_519>",
526
+ "<SPECIAL_520>",
527
+ "<SPECIAL_521>",
528
+ "<SPECIAL_522>",
529
+ "<SPECIAL_523>",
530
+ "<SPECIAL_524>",
531
+ "<SPECIAL_525>",
532
+ "<SPECIAL_526>",
533
+ "<SPECIAL_527>",
534
+ "<SPECIAL_528>",
535
+ "<SPECIAL_529>",
536
+ "<SPECIAL_530>",
537
+ "<SPECIAL_531>",
538
+ "<SPECIAL_532>",
539
+ "<SPECIAL_533>",
540
+ "<SPECIAL_534>",
541
+ "<SPECIAL_535>",
542
+ "<SPECIAL_536>",
543
+ "<SPECIAL_537>",
544
+ "<SPECIAL_538>",
545
+ "<SPECIAL_539>",
546
+ "<SPECIAL_540>",
547
+ "<SPECIAL_541>",
548
+ "<SPECIAL_542>",
549
+ "<SPECIAL_543>",
550
+ "<SPECIAL_544>",
551
+ "<SPECIAL_545>",
552
+ "<SPECIAL_546>",
553
+ "<SPECIAL_547>",
554
+ "<SPECIAL_548>",
555
+ "<SPECIAL_549>",
556
+ "<SPECIAL_550>",
557
+ "<SPECIAL_551>",
558
+ "<SPECIAL_552>",
559
+ "<SPECIAL_553>",
560
+ "<SPECIAL_554>",
561
+ "<SPECIAL_555>",
562
+ "<SPECIAL_556>",
563
+ "<SPECIAL_557>",
564
+ "<SPECIAL_558>",
565
+ "<SPECIAL_559>",
566
+ "<SPECIAL_560>",
567
+ "<SPECIAL_561>",
568
+ "<SPECIAL_562>",
569
+ "<SPECIAL_563>",
570
+ "<SPECIAL_564>",
571
+ "<SPECIAL_565>",
572
+ "<SPECIAL_566>",
573
+ "<SPECIAL_567>",
574
+ "<SPECIAL_568>",
575
+ "<SPECIAL_569>",
576
+ "<SPECIAL_570>",
577
+ "<SPECIAL_571>",
578
+ "<SPECIAL_572>",
579
+ "<SPECIAL_573>",
580
+ "<SPECIAL_574>",
581
+ "<SPECIAL_575>",
582
+ "<SPECIAL_576>",
583
+ "<SPECIAL_577>",
584
+ "<SPECIAL_578>",
585
+ "<SPECIAL_579>",
586
+ "<SPECIAL_580>",
587
+ "<SPECIAL_581>",
588
+ "<SPECIAL_582>",
589
+ "<SPECIAL_583>",
590
+ "<SPECIAL_584>",
591
+ "<SPECIAL_585>",
592
+ "<SPECIAL_586>",
593
+ "<SPECIAL_587>",
594
+ "<SPECIAL_588>",
595
+ "<SPECIAL_589>",
596
+ "<SPECIAL_590>",
597
+ "<SPECIAL_591>",
598
+ "<SPECIAL_592>",
599
+ "<SPECIAL_593>",
600
+ "<SPECIAL_594>",
601
+ "<SPECIAL_595>",
602
+ "<SPECIAL_596>",
603
+ "<SPECIAL_597>",
604
+ "<SPECIAL_598>",
605
+ "<SPECIAL_599>",
606
+ "<SPECIAL_600>",
607
+ "<SPECIAL_601>",
608
+ "<SPECIAL_602>",
609
+ "<SPECIAL_603>",
610
+ "<SPECIAL_604>",
611
+ "<SPECIAL_605>",
612
+ "<SPECIAL_606>",
613
+ "<SPECIAL_607>",
614
+ "<SPECIAL_608>",
615
+ "<SPECIAL_609>",
616
+ "<SPECIAL_610>",
617
+ "<SPECIAL_611>",
618
+ "<SPECIAL_612>",
619
+ "<SPECIAL_613>",
620
+ "<SPECIAL_614>",
621
+ "<SPECIAL_615>",
622
+ "<SPECIAL_616>",
623
+ "<SPECIAL_617>",
624
+ "<SPECIAL_618>",
625
+ "<SPECIAL_619>",
626
+ "<SPECIAL_620>",
627
+ "<SPECIAL_621>",
628
+ "<SPECIAL_622>",
629
+ "<SPECIAL_623>",
630
+ "<SPECIAL_624>",
631
+ "<SPECIAL_625>",
632
+ "<SPECIAL_626>",
633
+ "<SPECIAL_627>",
634
+ "<SPECIAL_628>",
635
+ "<SPECIAL_629>",
636
+ "<SPECIAL_630>",
637
+ "<SPECIAL_631>",
638
+ "<SPECIAL_632>",
639
+ "<SPECIAL_633>",
640
+ "<SPECIAL_634>",
641
+ "<SPECIAL_635>",
642
+ "<SPECIAL_636>",
643
+ "<SPECIAL_637>",
644
+ "<SPECIAL_638>",
645
+ "<SPECIAL_639>",
646
+ "<SPECIAL_640>",
647
+ "<SPECIAL_641>",
648
+ "<SPECIAL_642>",
649
+ "<SPECIAL_643>",
650
+ "<SPECIAL_644>",
651
+ "<SPECIAL_645>",
652
+ "<SPECIAL_646>",
653
+ "<SPECIAL_647>",
654
+ "<SPECIAL_648>",
655
+ "<SPECIAL_649>",
656
+ "<SPECIAL_650>",
657
+ "<SPECIAL_651>",
658
+ "<SPECIAL_652>",
659
+ "<SPECIAL_653>",
660
+ "<SPECIAL_654>",
661
+ "<SPECIAL_655>",
662
+ "<SPECIAL_656>",
663
+ "<SPECIAL_657>",
664
+ "<SPECIAL_658>",
665
+ "<SPECIAL_659>",
666
+ "<SPECIAL_660>",
667
+ "<SPECIAL_661>",
668
+ "<SPECIAL_662>",
669
+ "<SPECIAL_663>",
670
+ "<SPECIAL_664>",
671
+ "<SPECIAL_665>",
672
+ "<SPECIAL_666>",
673
+ "<SPECIAL_667>",
674
+ "<SPECIAL_668>",
675
+ "<SPECIAL_669>",
676
+ "<SPECIAL_670>",
677
+ "<SPECIAL_671>",
678
+ "<SPECIAL_672>",
679
+ "<SPECIAL_673>",
680
+ "<SPECIAL_674>",
681
+ "<SPECIAL_675>",
682
+ "<SPECIAL_676>",
683
+ "<SPECIAL_677>",
684
+ "<SPECIAL_678>",
685
+ "<SPECIAL_679>",
686
+ "<SPECIAL_680>",
687
+ "<SPECIAL_681>",
688
+ "<SPECIAL_682>",
689
+ "<SPECIAL_683>",
690
+ "<SPECIAL_684>",
691
+ "<SPECIAL_685>",
692
+ "<SPECIAL_686>",
693
+ "<SPECIAL_687>",
694
+ "<SPECIAL_688>",
695
+ "<SPECIAL_689>",
696
+ "<SPECIAL_690>",
697
+ "<SPECIAL_691>",
698
+ "<SPECIAL_692>",
699
+ "<SPECIAL_693>",
700
+ "<SPECIAL_694>",
701
+ "<SPECIAL_695>",
702
+ "<SPECIAL_696>",
703
+ "<SPECIAL_697>",
704
+ "<SPECIAL_698>",
705
+ "<SPECIAL_699>",
706
+ "<SPECIAL_700>",
707
+ "<SPECIAL_701>",
708
+ "<SPECIAL_702>",
709
+ "<SPECIAL_703>",
710
+ "<SPECIAL_704>",
711
+ "<SPECIAL_705>",
712
+ "<SPECIAL_706>",
713
+ "<SPECIAL_707>",
714
+ "<SPECIAL_708>",
715
+ "<SPECIAL_709>",
716
+ "<SPECIAL_710>",
717
+ "<SPECIAL_711>",
718
+ "<SPECIAL_712>",
719
+ "<SPECIAL_713>",
720
+ "<SPECIAL_714>",
721
+ "<SPECIAL_715>",
722
+ "<SPECIAL_716>",
723
+ "<SPECIAL_717>",
724
+ "<SPECIAL_718>",
725
+ "<SPECIAL_719>",
726
+ "<SPECIAL_720>",
727
+ "<SPECIAL_721>",
728
+ "<SPECIAL_722>",
729
+ "<SPECIAL_723>",
730
+ "<SPECIAL_724>",
731
+ "<SPECIAL_725>",
732
+ "<SPECIAL_726>",
733
+ "<SPECIAL_727>",
734
+ "<SPECIAL_728>",
735
+ "<SPECIAL_729>",
736
+ "<SPECIAL_730>",
737
+ "<SPECIAL_731>",
738
+ "<SPECIAL_732>",
739
+ "<SPECIAL_733>",
740
+ "<SPECIAL_734>",
741
+ "<SPECIAL_735>",
742
+ "<SPECIAL_736>",
743
+ "<SPECIAL_737>",
744
+ "<SPECIAL_738>",
745
+ "<SPECIAL_739>",
746
+ "<SPECIAL_740>",
747
+ "<SPECIAL_741>",
748
+ "<SPECIAL_742>",
749
+ "<SPECIAL_743>",
750
+ "<SPECIAL_744>",
751
+ "<SPECIAL_745>",
752
+ "<SPECIAL_746>",
753
+ "<SPECIAL_747>",
754
+ "<SPECIAL_748>",
755
+ "<SPECIAL_749>",
756
+ "<SPECIAL_750>",
757
+ "<SPECIAL_751>",
758
+ "<SPECIAL_752>",
759
+ "<SPECIAL_753>",
760
+ "<SPECIAL_754>",
761
+ "<SPECIAL_755>",
762
+ "<SPECIAL_756>",
763
+ "<SPECIAL_757>",
764
+ "<SPECIAL_758>",
765
+ "<SPECIAL_759>",
766
+ "<SPECIAL_760>",
767
+ "<SPECIAL_761>",
768
+ "<SPECIAL_762>",
769
+ "<SPECIAL_763>",
770
+ "<SPECIAL_764>",
771
+ "<SPECIAL_765>",
772
+ "<SPECIAL_766>",
773
+ "<SPECIAL_767>",
774
+ "<SPECIAL_768>",
775
+ "<SPECIAL_769>",
776
+ "<SPECIAL_770>",
777
+ "<SPECIAL_771>",
778
+ "<SPECIAL_772>",
779
+ "<SPECIAL_773>",
780
+ "<SPECIAL_774>",
781
+ "<SPECIAL_775>",
782
+ "<SPECIAL_776>",
783
+ "<SPECIAL_777>",
784
+ "<SPECIAL_778>",
785
+ "<SPECIAL_779>",
786
+ "<SPECIAL_780>",
787
+ "<SPECIAL_781>",
788
+ "<SPECIAL_782>",
789
+ "<SPECIAL_783>",
790
+ "<SPECIAL_784>",
791
+ "<SPECIAL_785>",
792
+ "<SPECIAL_786>",
793
+ "<SPECIAL_787>",
794
+ "<SPECIAL_788>",
795
+ "<SPECIAL_789>",
796
+ "<SPECIAL_790>",
797
+ "<SPECIAL_791>",
798
+ "<SPECIAL_792>",
799
+ "<SPECIAL_793>",
800
+ "<SPECIAL_794>",
801
+ "<SPECIAL_795>",
802
+ "<SPECIAL_796>",
803
+ "<SPECIAL_797>",
804
+ "<SPECIAL_798>",
805
+ "<SPECIAL_799>",
806
+ "<SPECIAL_800>",
807
+ "<SPECIAL_801>",
808
+ "<SPECIAL_802>",
809
+ "<SPECIAL_803>",
810
+ "<SPECIAL_804>",
811
+ "<SPECIAL_805>",
812
+ "<SPECIAL_806>",
813
+ "<SPECIAL_807>",
814
+ "<SPECIAL_808>",
815
+ "<SPECIAL_809>",
816
+ "<SPECIAL_810>",
817
+ "<SPECIAL_811>",
818
+ "<SPECIAL_812>",
819
+ "<SPECIAL_813>",
820
+ "<SPECIAL_814>",
821
+ "<SPECIAL_815>",
822
+ "<SPECIAL_816>",
823
+ "<SPECIAL_817>",
824
+ "<SPECIAL_818>",
825
+ "<SPECIAL_819>",
826
+ "<SPECIAL_820>",
827
+ "<SPECIAL_821>",
828
+ "<SPECIAL_822>",
829
+ "<SPECIAL_823>",
830
+ "<SPECIAL_824>",
831
+ "<SPECIAL_825>",
832
+ "<SPECIAL_826>",
833
+ "<SPECIAL_827>",
834
+ "<SPECIAL_828>",
835
+ "<SPECIAL_829>",
836
+ "<SPECIAL_830>",
837
+ "<SPECIAL_831>",
838
+ "<SPECIAL_832>",
839
+ "<SPECIAL_833>",
840
+ "<SPECIAL_834>",
841
+ "<SPECIAL_835>",
842
+ "<SPECIAL_836>",
843
+ "<SPECIAL_837>",
844
+ "<SPECIAL_838>",
845
+ "<SPECIAL_839>",
846
+ "<SPECIAL_840>",
847
+ "<SPECIAL_841>",
848
+ "<SPECIAL_842>",
849
+ "<SPECIAL_843>",
850
+ "<SPECIAL_844>",
851
+ "<SPECIAL_845>",
852
+ "<SPECIAL_846>",
853
+ "<SPECIAL_847>",
854
+ "<SPECIAL_848>",
855
+ "<SPECIAL_849>",
856
+ "<SPECIAL_850>",
857
+ "<SPECIAL_851>",
858
+ "<SPECIAL_852>",
859
+ "<SPECIAL_853>",
860
+ "<SPECIAL_854>",
861
+ "<SPECIAL_855>",
862
+ "<SPECIAL_856>",
863
+ "<SPECIAL_857>",
864
+ "<SPECIAL_858>",
865
+ "<SPECIAL_859>",
866
+ "<SPECIAL_860>",
867
+ "<SPECIAL_861>",
868
+ "<SPECIAL_862>",
869
+ "<SPECIAL_863>",
870
+ "<SPECIAL_864>",
871
+ "<SPECIAL_865>",
872
+ "<SPECIAL_866>",
873
+ "<SPECIAL_867>",
874
+ "<SPECIAL_868>",
875
+ "<SPECIAL_869>",
876
+ "<SPECIAL_870>",
877
+ "<SPECIAL_871>",
878
+ "<SPECIAL_872>",
879
+ "<SPECIAL_873>",
880
+ "<SPECIAL_874>",
881
+ "<SPECIAL_875>",
882
+ "<SPECIAL_876>",
883
+ "<SPECIAL_877>",
884
+ "<SPECIAL_878>",
885
+ "<SPECIAL_879>",
886
+ "<SPECIAL_880>",
887
+ "<SPECIAL_881>",
888
+ "<SPECIAL_882>",
889
+ "<SPECIAL_883>",
890
+ "<SPECIAL_884>",
891
+ "<SPECIAL_885>",
892
+ "<SPECIAL_886>",
893
+ "<SPECIAL_887>",
894
+ "<SPECIAL_888>",
895
+ "<SPECIAL_889>",
896
+ "<SPECIAL_890>",
897
+ "<SPECIAL_891>",
898
+ "<SPECIAL_892>",
899
+ "<SPECIAL_893>",
900
+ "<SPECIAL_894>",
901
+ "<SPECIAL_895>",
902
+ "<SPECIAL_896>",
903
+ "<SPECIAL_897>",
904
+ "<SPECIAL_898>",
905
+ "<SPECIAL_899>",
906
+ "<SPECIAL_900>",
907
+ "<SPECIAL_901>",
908
+ "<SPECIAL_902>",
909
+ "<SPECIAL_903>",
910
+ "<SPECIAL_904>",
911
+ "<SPECIAL_905>",
912
+ "<SPECIAL_906>",
913
+ "<SPECIAL_907>",
914
+ "<SPECIAL_908>",
915
+ "<SPECIAL_909>",
916
+ "<SPECIAL_910>",
917
+ "<SPECIAL_911>",
918
+ "<SPECIAL_912>",
919
+ "<SPECIAL_913>",
920
+ "<SPECIAL_914>",
921
+ "<SPECIAL_915>",
922
+ "<SPECIAL_916>",
923
+ "<SPECIAL_917>",
924
+ "<SPECIAL_918>",
925
+ "<SPECIAL_919>",
926
+ "<SPECIAL_920>",
927
+ "<SPECIAL_921>",
928
+ "<SPECIAL_922>",
929
+ "<SPECIAL_923>",
930
+ "<SPECIAL_924>",
931
+ "<SPECIAL_925>",
932
+ "<SPECIAL_926>",
933
+ "<SPECIAL_927>",
934
+ "<SPECIAL_928>",
935
+ "<SPECIAL_929>",
936
+ "<SPECIAL_930>",
937
+ "<SPECIAL_931>",
938
+ "<SPECIAL_932>",
939
+ "<SPECIAL_933>",
940
+ "<SPECIAL_934>",
941
+ "<SPECIAL_935>",
942
+ "<SPECIAL_936>",
943
+ "<SPECIAL_937>",
944
+ "<SPECIAL_938>",
945
+ "<SPECIAL_939>",
946
+ "<SPECIAL_940>",
947
+ "<SPECIAL_941>",
948
+ "<SPECIAL_942>",
949
+ "<SPECIAL_943>",
950
+ "<SPECIAL_944>",
951
+ "<SPECIAL_945>",
952
+ "<SPECIAL_946>",
953
+ "<SPECIAL_947>",
954
+ "<SPECIAL_948>",
955
+ "<SPECIAL_949>",
956
+ "<SPECIAL_950>",
957
+ "<SPECIAL_951>",
958
+ "<SPECIAL_952>",
959
+ "<SPECIAL_953>",
960
+ "<SPECIAL_954>",
961
+ "<SPECIAL_955>",
962
+ "<SPECIAL_956>",
963
+ "<SPECIAL_957>",
964
+ "<SPECIAL_958>",
965
+ "<SPECIAL_959>",
966
+ "<SPECIAL_960>",
967
+ "<SPECIAL_961>",
968
+ "<SPECIAL_962>",
969
+ "<SPECIAL_963>",
970
+ "<SPECIAL_964>",
971
+ "<SPECIAL_965>",
972
+ "<SPECIAL_966>",
973
+ "<SPECIAL_967>",
974
+ "<SPECIAL_968>",
975
+ "<SPECIAL_969>",
976
+ "<SPECIAL_970>",
977
+ "<SPECIAL_971>",
978
+ "<SPECIAL_972>",
979
+ "<SPECIAL_973>",
980
+ "<SPECIAL_974>",
981
+ "<SPECIAL_975>",
982
+ "<SPECIAL_976>",
983
+ "<SPECIAL_977>",
984
+ "<SPECIAL_978>",
985
+ "<SPECIAL_979>",
986
+ "<SPECIAL_980>",
987
+ "<SPECIAL_981>",
988
+ "<SPECIAL_982>",
989
+ "<SPECIAL_983>",
990
+ "<SPECIAL_984>",
991
+ "<SPECIAL_985>",
992
+ "<SPECIAL_986>",
993
+ "<SPECIAL_987>",
994
+ "<SPECIAL_988>",
995
+ "<SPECIAL_989>",
996
+ "<SPECIAL_990>",
997
+ "<SPECIAL_991>",
998
+ "<SPECIAL_992>",
999
+ "<SPECIAL_993>",
1000
+ "<SPECIAL_994>",
1001
+ "<SPECIAL_995>",
1002
+ "<SPECIAL_996>",
1003
+ "<SPECIAL_997>",
1004
+ "<SPECIAL_998>",
1005
+ "<SPECIAL_999>"
1006
+ ],
1007
+ "is_local": true,
1008
+ "model_max_length": 2048,
1009
+ "pad_token": "<pad>",
1010
+ "processor_class": "PixtralProcessor",
1011
+ "tokenizer_class": "TokenizersBackend",
1012
+ "unk_token": "<unk>"
1013
+ }
scheduler/scheduler_config.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "FlowMatchEulerDiscreteScheduler",
3
+ "_diffusers_version": "0.36.0",
4
+ "base_image_seq_len": 256,
5
+ "base_shift": 0.5,
6
+ "invert_sigmas": false,
7
+ "max_image_seq_len": 4096,
8
+ "max_shift": 1.15,
9
+ "num_train_timesteps": 1000,
10
+ "shift": 3.0,
11
+ "shift_terminal": null,
12
+ "stochastic_sampling": false,
13
+ "time_shift_type": "exponential",
14
+ "use_beta_sigmas": false,
15
+ "use_dynamic_shifting": false,
16
+ "use_exponential_sigmas": false,
17
+ "use_karras_sigmas": false
18
+ }
text_encoder/config.json ADDED
@@ -0,0 +1,480 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Mistral3Model"
4
+ ],
5
+ "dtype": "bfloat16",
6
+ "image_token_index": 10,
7
+ "model_type": "mistral3",
8
+ "multimodal_projector_bias": false,
9
+ "projector_hidden_act": "gelu",
10
+ "quantization_config": {
11
+ "add_skip_keys": true,
12
+ "dequantize_fp32": false,
13
+ "dynamic_loss_threshold": 0.0001,
14
+ "group_size": 0,
15
+ "is_integer": true,
16
+ "is_training": false,
17
+ "modules_dtype_dict": {
18
+ "float10_e4m5fn": [
19
+ "vision_tower.transformer.layers.0.attention.k_proj.weight",
20
+ "vision_tower.transformer.layers.0.attention.q_proj.weight"
21
+ ],
22
+ "float9_e3m5fn": [
23
+ "vision_tower.transformer.layers.0.attention.o_proj.weight",
24
+ "vision_tower.transformer.layers.0.attention.v_proj.weight",
25
+ "vision_tower.transformer.layers.0.feed_forward.gate_proj.weight",
26
+ "vision_tower.transformer.layers.0.feed_forward.up_proj.weight",
27
+ "vision_tower.transformer.layers.1.attention.k_proj.weight",
28
+ "vision_tower.transformer.layers.1.attention.o_proj.weight",
29
+ "vision_tower.transformer.layers.1.attention.q_proj.weight",
30
+ "vision_tower.transformer.layers.1.feed_forward.gate_proj.weight",
31
+ "vision_tower.transformer.layers.2.attention.o_proj.weight",
32
+ "vision_tower.transformer.layers.23.feed_forward.down_proj.weight"
33
+ ],
34
+ "int8": [
35
+ "language_model.layers.0.mlp.down_proj.weight",
36
+ "language_model.layers.0.mlp.gate_proj.weight",
37
+ "language_model.layers.0.mlp.up_proj.weight",
38
+ "language_model.layers.0.self_attn.k_proj.weight",
39
+ "language_model.layers.0.self_attn.q_proj.weight",
40
+ "language_model.layers.0.self_attn.v_proj.weight",
41
+ "language_model.layers.1.mlp.down_proj.weight",
42
+ "language_model.layers.1.mlp.gate_proj.weight",
43
+ "language_model.layers.1.mlp.up_proj.weight",
44
+ "language_model.layers.1.self_attn.k_proj.weight",
45
+ "language_model.layers.1.self_attn.o_proj.weight",
46
+ "language_model.layers.1.self_attn.q_proj.weight",
47
+ "language_model.layers.1.self_attn.v_proj.weight",
48
+ "language_model.layers.2.mlp.down_proj.weight",
49
+ "language_model.layers.2.mlp.gate_proj.weight",
50
+ "language_model.layers.2.mlp.up_proj.weight",
51
+ "language_model.layers.2.self_attn.k_proj.weight",
52
+ "language_model.layers.2.self_attn.o_proj.weight",
53
+ "language_model.layers.2.self_attn.q_proj.weight",
54
+ "language_model.layers.2.self_attn.v_proj.weight",
55
+ "language_model.layers.3.mlp.down_proj.weight",
56
+ "language_model.layers.3.mlp.gate_proj.weight",
57
+ "language_model.layers.3.mlp.up_proj.weight",
58
+ "language_model.layers.3.self_attn.k_proj.weight",
59
+ "language_model.layers.3.self_attn.o_proj.weight",
60
+ "language_model.layers.3.self_attn.q_proj.weight",
61
+ "language_model.layers.3.self_attn.v_proj.weight",
62
+ "language_model.layers.4.mlp.down_proj.weight",
63
+ "language_model.layers.4.mlp.gate_proj.weight",
64
+ "language_model.layers.4.mlp.up_proj.weight",
65
+ "language_model.layers.4.self_attn.k_proj.weight",
66
+ "language_model.layers.4.self_attn.o_proj.weight",
67
+ "language_model.layers.4.self_attn.q_proj.weight",
68
+ "language_model.layers.4.self_attn.v_proj.weight",
69
+ "language_model.layers.5.mlp.down_proj.weight",
70
+ "language_model.layers.5.mlp.gate_proj.weight",
71
+ "language_model.layers.5.mlp.up_proj.weight",
72
+ "language_model.layers.5.self_attn.k_proj.weight",
73
+ "language_model.layers.5.self_attn.o_proj.weight",
74
+ "language_model.layers.5.self_attn.q_proj.weight",
75
+ "language_model.layers.5.self_attn.v_proj.weight",
76
+ "language_model.layers.6.mlp.down_proj.weight",
77
+ "language_model.layers.6.mlp.gate_proj.weight",
78
+ "language_model.layers.6.mlp.up_proj.weight",
79
+ "language_model.layers.6.self_attn.k_proj.weight",
80
+ "language_model.layers.6.self_attn.o_proj.weight",
81
+ "language_model.layers.6.self_attn.q_proj.weight",
82
+ "language_model.layers.6.self_attn.v_proj.weight",
83
+ "language_model.layers.7.mlp.down_proj.weight",
84
+ "language_model.layers.7.mlp.gate_proj.weight",
85
+ "language_model.layers.7.mlp.up_proj.weight",
86
+ "language_model.layers.7.self_attn.k_proj.weight",
87
+ "language_model.layers.7.self_attn.o_proj.weight",
88
+ "language_model.layers.7.self_attn.q_proj.weight",
89
+ "language_model.layers.7.self_attn.v_proj.weight",
90
+ "language_model.layers.8.mlp.down_proj.weight",
91
+ "language_model.layers.8.mlp.gate_proj.weight",
92
+ "language_model.layers.8.mlp.up_proj.weight",
93
+ "language_model.layers.8.self_attn.k_proj.weight",
94
+ "language_model.layers.8.self_attn.o_proj.weight",
95
+ "language_model.layers.8.self_attn.q_proj.weight",
96
+ "language_model.layers.8.self_attn.v_proj.weight",
97
+ "language_model.layers.9.mlp.down_proj.weight",
98
+ "language_model.layers.9.mlp.gate_proj.weight",
99
+ "language_model.layers.9.mlp.up_proj.weight",
100
+ "language_model.layers.9.self_attn.k_proj.weight",
101
+ "language_model.layers.9.self_attn.o_proj.weight",
102
+ "language_model.layers.9.self_attn.q_proj.weight",
103
+ "language_model.layers.9.self_attn.v_proj.weight",
104
+ "language_model.layers.10.mlp.down_proj.weight",
105
+ "language_model.layers.10.mlp.gate_proj.weight",
106
+ "language_model.layers.10.mlp.up_proj.weight",
107
+ "language_model.layers.10.self_attn.k_proj.weight",
108
+ "language_model.layers.10.self_attn.o_proj.weight",
109
+ "language_model.layers.10.self_attn.q_proj.weight",
110
+ "language_model.layers.10.self_attn.v_proj.weight",
111
+ "language_model.layers.11.mlp.down_proj.weight",
112
+ "language_model.layers.11.mlp.gate_proj.weight",
113
+ "language_model.layers.11.mlp.up_proj.weight",
114
+ "language_model.layers.11.self_attn.k_proj.weight",
115
+ "language_model.layers.11.self_attn.o_proj.weight",
116
+ "language_model.layers.11.self_attn.q_proj.weight",
117
+ "language_model.layers.11.self_attn.v_proj.weight",
118
+ "language_model.layers.12.mlp.down_proj.weight",
119
+ "language_model.layers.12.mlp.gate_proj.weight",
120
+ "language_model.layers.12.mlp.up_proj.weight",
121
+ "language_model.layers.12.self_attn.k_proj.weight",
122
+ "language_model.layers.12.self_attn.o_proj.weight",
123
+ "language_model.layers.12.self_attn.q_proj.weight",
124
+ "language_model.layers.12.self_attn.v_proj.weight",
125
+ "language_model.layers.13.mlp.down_proj.weight",
126
+ "language_model.layers.13.mlp.gate_proj.weight",
127
+ "language_model.layers.13.mlp.up_proj.weight",
128
+ "language_model.layers.13.self_attn.k_proj.weight",
129
+ "language_model.layers.13.self_attn.o_proj.weight",
130
+ "language_model.layers.13.self_attn.q_proj.weight",
131
+ "language_model.layers.13.self_attn.v_proj.weight",
132
+ "language_model.layers.14.mlp.down_proj.weight",
133
+ "language_model.layers.14.mlp.gate_proj.weight",
134
+ "language_model.layers.14.mlp.up_proj.weight",
135
+ "language_model.layers.14.self_attn.k_proj.weight",
136
+ "language_model.layers.14.self_attn.o_proj.weight",
137
+ "language_model.layers.14.self_attn.q_proj.weight",
138
+ "language_model.layers.14.self_attn.v_proj.weight",
139
+ "language_model.layers.15.mlp.down_proj.weight",
140
+ "language_model.layers.15.mlp.gate_proj.weight",
141
+ "language_model.layers.15.mlp.up_proj.weight",
142
+ "language_model.layers.15.self_attn.k_proj.weight",
143
+ "language_model.layers.15.self_attn.o_proj.weight",
144
+ "language_model.layers.15.self_attn.q_proj.weight",
145
+ "language_model.layers.15.self_attn.v_proj.weight",
146
+ "language_model.layers.16.mlp.down_proj.weight",
147
+ "language_model.layers.16.mlp.gate_proj.weight",
148
+ "language_model.layers.16.mlp.up_proj.weight",
149
+ "language_model.layers.16.self_attn.k_proj.weight",
150
+ "language_model.layers.16.self_attn.o_proj.weight",
151
+ "language_model.layers.16.self_attn.q_proj.weight",
152
+ "language_model.layers.16.self_attn.v_proj.weight",
153
+ "language_model.layers.17.mlp.down_proj.weight",
154
+ "language_model.layers.17.mlp.gate_proj.weight",
155
+ "language_model.layers.17.mlp.up_proj.weight",
156
+ "language_model.layers.17.self_attn.k_proj.weight",
157
+ "language_model.layers.17.self_attn.o_proj.weight",
158
+ "language_model.layers.17.self_attn.q_proj.weight",
159
+ "language_model.layers.17.self_attn.v_proj.weight",
160
+ "language_model.layers.18.mlp.down_proj.weight",
161
+ "language_model.layers.18.mlp.gate_proj.weight",
162
+ "language_model.layers.18.mlp.up_proj.weight",
163
+ "language_model.layers.18.self_attn.k_proj.weight",
164
+ "language_model.layers.18.self_attn.o_proj.weight",
165
+ "language_model.layers.18.self_attn.q_proj.weight",
166
+ "language_model.layers.18.self_attn.v_proj.weight",
167
+ "language_model.layers.19.mlp.down_proj.weight",
168
+ "language_model.layers.19.mlp.gate_proj.weight",
169
+ "language_model.layers.19.mlp.up_proj.weight",
170
+ "language_model.layers.19.self_attn.k_proj.weight",
171
+ "language_model.layers.19.self_attn.o_proj.weight",
172
+ "language_model.layers.19.self_attn.q_proj.weight",
173
+ "language_model.layers.19.self_attn.v_proj.weight",
174
+ "language_model.layers.20.mlp.down_proj.weight",
175
+ "language_model.layers.20.mlp.gate_proj.weight",
176
+ "language_model.layers.20.mlp.up_proj.weight",
177
+ "language_model.layers.20.self_attn.k_proj.weight",
178
+ "language_model.layers.20.self_attn.o_proj.weight",
179
+ "language_model.layers.20.self_attn.q_proj.weight",
180
+ "language_model.layers.20.self_attn.v_proj.weight",
181
+ "language_model.layers.21.mlp.down_proj.weight",
182
+ "language_model.layers.21.mlp.gate_proj.weight",
183
+ "language_model.layers.21.mlp.up_proj.weight",
184
+ "language_model.layers.21.self_attn.k_proj.weight",
185
+ "language_model.layers.21.self_attn.o_proj.weight",
186
+ "language_model.layers.21.self_attn.q_proj.weight",
187
+ "language_model.layers.21.self_attn.v_proj.weight",
188
+ "language_model.layers.22.mlp.down_proj.weight",
189
+ "language_model.layers.22.mlp.gate_proj.weight",
190
+ "language_model.layers.22.mlp.up_proj.weight",
191
+ "language_model.layers.22.self_attn.k_proj.weight",
192
+ "language_model.layers.22.self_attn.o_proj.weight",
193
+ "language_model.layers.22.self_attn.q_proj.weight",
194
+ "language_model.layers.22.self_attn.v_proj.weight",
195
+ "language_model.layers.23.mlp.down_proj.weight",
196
+ "language_model.layers.23.mlp.gate_proj.weight",
197
+ "language_model.layers.23.mlp.up_proj.weight",
198
+ "language_model.layers.23.self_attn.k_proj.weight",
199
+ "language_model.layers.23.self_attn.o_proj.weight",
200
+ "language_model.layers.23.self_attn.q_proj.weight",
201
+ "language_model.layers.23.self_attn.v_proj.weight",
202
+ "language_model.layers.24.mlp.down_proj.weight",
203
+ "language_model.layers.24.mlp.gate_proj.weight",
204
+ "language_model.layers.24.mlp.up_proj.weight",
205
+ "language_model.layers.24.self_attn.k_proj.weight",
206
+ "language_model.layers.24.self_attn.o_proj.weight",
207
+ "language_model.layers.24.self_attn.q_proj.weight",
208
+ "language_model.layers.24.self_attn.v_proj.weight",
209
+ "language_model.layers.25.mlp.down_proj.weight",
210
+ "language_model.layers.25.mlp.gate_proj.weight",
211
+ "language_model.layers.25.mlp.up_proj.weight",
212
+ "language_model.layers.25.self_attn.k_proj.weight",
213
+ "language_model.layers.25.self_attn.o_proj.weight",
214
+ "language_model.layers.25.self_attn.q_proj.weight",
215
+ "language_model.layers.25.self_attn.v_proj.weight",
216
+ "vision_tower.transformer.layers.0.feed_forward.down_proj.weight",
217
+ "vision_tower.transformer.layers.1.feed_forward.up_proj.weight",
218
+ "vision_tower.transformer.layers.2.attention.k_proj.weight",
219
+ "vision_tower.transformer.layers.2.attention.q_proj.weight",
220
+ "vision_tower.transformer.layers.2.attention.v_proj.weight",
221
+ "vision_tower.transformer.layers.2.feed_forward.gate_proj.weight",
222
+ "vision_tower.transformer.layers.2.feed_forward.up_proj.weight",
223
+ "vision_tower.transformer.layers.3.attention.q_proj.weight",
224
+ "vision_tower.transformer.layers.3.attention.v_proj.weight",
225
+ "vision_tower.transformer.layers.3.feed_forward.down_proj.weight",
226
+ "vision_tower.transformer.layers.3.feed_forward.gate_proj.weight",
227
+ "vision_tower.transformer.layers.3.feed_forward.up_proj.weight",
228
+ "vision_tower.transformer.layers.4.attention.q_proj.weight",
229
+ "vision_tower.transformer.layers.4.attention.v_proj.weight",
230
+ "vision_tower.transformer.layers.4.feed_forward.down_proj.weight",
231
+ "vision_tower.transformer.layers.4.feed_forward.gate_proj.weight",
232
+ "vision_tower.transformer.layers.4.feed_forward.up_proj.weight",
233
+ "vision_tower.transformer.layers.5.attention.o_proj.weight",
234
+ "vision_tower.transformer.layers.5.attention.q_proj.weight",
235
+ "vision_tower.transformer.layers.5.attention.v_proj.weight",
236
+ "vision_tower.transformer.layers.5.feed_forward.down_proj.weight",
237
+ "vision_tower.transformer.layers.5.feed_forward.gate_proj.weight",
238
+ "vision_tower.transformer.layers.5.feed_forward.up_proj.weight",
239
+ "vision_tower.transformer.layers.6.attention.k_proj.weight",
240
+ "vision_tower.transformer.layers.6.attention.o_proj.weight",
241
+ "vision_tower.transformer.layers.6.attention.q_proj.weight",
242
+ "vision_tower.transformer.layers.6.attention.v_proj.weight",
243
+ "vision_tower.transformer.layers.6.feed_forward.down_proj.weight",
244
+ "vision_tower.transformer.layers.6.feed_forward.gate_proj.weight",
245
+ "vision_tower.transformer.layers.6.feed_forward.up_proj.weight",
246
+ "vision_tower.transformer.layers.7.attention.k_proj.weight",
247
+ "vision_tower.transformer.layers.7.attention.o_proj.weight",
248
+ "vision_tower.transformer.layers.7.attention.q_proj.weight",
249
+ "vision_tower.transformer.layers.7.attention.v_proj.weight",
250
+ "vision_tower.transformer.layers.7.feed_forward.down_proj.weight",
251
+ "vision_tower.transformer.layers.7.feed_forward.gate_proj.weight",
252
+ "vision_tower.transformer.layers.7.feed_forward.up_proj.weight",
253
+ "vision_tower.transformer.layers.8.attention.k_proj.weight",
254
+ "vision_tower.transformer.layers.8.attention.o_proj.weight",
255
+ "vision_tower.transformer.layers.8.attention.q_proj.weight",
256
+ "vision_tower.transformer.layers.8.attention.v_proj.weight",
257
+ "vision_tower.transformer.layers.8.feed_forward.down_proj.weight",
258
+ "vision_tower.transformer.layers.8.feed_forward.gate_proj.weight",
259
+ "vision_tower.transformer.layers.8.feed_forward.up_proj.weight",
260
+ "vision_tower.transformer.layers.9.attention.k_proj.weight",
261
+ "vision_tower.transformer.layers.9.attention.o_proj.weight",
262
+ "vision_tower.transformer.layers.9.attention.q_proj.weight",
263
+ "vision_tower.transformer.layers.9.attention.v_proj.weight",
264
+ "vision_tower.transformer.layers.9.feed_forward.down_proj.weight",
265
+ "vision_tower.transformer.layers.9.feed_forward.gate_proj.weight",
266
+ "vision_tower.transformer.layers.9.feed_forward.up_proj.weight",
267
+ "vision_tower.transformer.layers.10.attention.k_proj.weight",
268
+ "vision_tower.transformer.layers.10.attention.o_proj.weight",
269
+ "vision_tower.transformer.layers.10.attention.q_proj.weight",
270
+ "vision_tower.transformer.layers.10.attention.v_proj.weight",
271
+ "vision_tower.transformer.layers.10.feed_forward.down_proj.weight",
272
+ "vision_tower.transformer.layers.10.feed_forward.gate_proj.weight",
273
+ "vision_tower.transformer.layers.10.feed_forward.up_proj.weight",
274
+ "vision_tower.transformer.layers.11.attention.k_proj.weight",
275
+ "vision_tower.transformer.layers.11.attention.o_proj.weight",
276
+ "vision_tower.transformer.layers.11.attention.q_proj.weight",
277
+ "vision_tower.transformer.layers.11.attention.v_proj.weight",
278
+ "vision_tower.transformer.layers.11.feed_forward.down_proj.weight",
279
+ "vision_tower.transformer.layers.11.feed_forward.gate_proj.weight",
280
+ "vision_tower.transformer.layers.11.feed_forward.up_proj.weight",
281
+ "vision_tower.transformer.layers.12.attention.k_proj.weight",
282
+ "vision_tower.transformer.layers.12.attention.o_proj.weight",
283
+ "vision_tower.transformer.layers.12.attention.q_proj.weight",
284
+ "vision_tower.transformer.layers.12.attention.v_proj.weight",
285
+ "vision_tower.transformer.layers.12.feed_forward.down_proj.weight",
286
+ "vision_tower.transformer.layers.12.feed_forward.gate_proj.weight",
287
+ "vision_tower.transformer.layers.12.feed_forward.up_proj.weight",
288
+ "vision_tower.transformer.layers.13.attention.k_proj.weight",
289
+ "vision_tower.transformer.layers.13.attention.o_proj.weight",
290
+ "vision_tower.transformer.layers.13.attention.q_proj.weight",
291
+ "vision_tower.transformer.layers.13.attention.v_proj.weight",
292
+ "vision_tower.transformer.layers.13.feed_forward.down_proj.weight",
293
+ "vision_tower.transformer.layers.13.feed_forward.gate_proj.weight",
294
+ "vision_tower.transformer.layers.13.feed_forward.up_proj.weight",
295
+ "vision_tower.transformer.layers.14.attention.k_proj.weight",
296
+ "vision_tower.transformer.layers.14.attention.o_proj.weight",
297
+ "vision_tower.transformer.layers.14.attention.q_proj.weight",
298
+ "vision_tower.transformer.layers.14.attention.v_proj.weight",
299
+ "vision_tower.transformer.layers.14.feed_forward.down_proj.weight",
300
+ "vision_tower.transformer.layers.14.feed_forward.gate_proj.weight",
301
+ "vision_tower.transformer.layers.14.feed_forward.up_proj.weight",
302
+ "vision_tower.transformer.layers.15.attention.k_proj.weight",
303
+ "vision_tower.transformer.layers.15.attention.o_proj.weight",
304
+ "vision_tower.transformer.layers.15.attention.q_proj.weight",
305
+ "vision_tower.transformer.layers.15.attention.v_proj.weight",
306
+ "vision_tower.transformer.layers.15.feed_forward.down_proj.weight",
307
+ "vision_tower.transformer.layers.15.feed_forward.gate_proj.weight",
308
+ "vision_tower.transformer.layers.15.feed_forward.up_proj.weight",
309
+ "vision_tower.transformer.layers.16.attention.k_proj.weight",
310
+ "vision_tower.transformer.layers.16.attention.o_proj.weight",
311
+ "vision_tower.transformer.layers.16.attention.q_proj.weight",
312
+ "vision_tower.transformer.layers.16.attention.v_proj.weight",
313
+ "vision_tower.transformer.layers.16.feed_forward.down_proj.weight",
314
+ "vision_tower.transformer.layers.16.feed_forward.gate_proj.weight",
315
+ "vision_tower.transformer.layers.16.feed_forward.up_proj.weight",
316
+ "vision_tower.transformer.layers.17.attention.k_proj.weight",
317
+ "vision_tower.transformer.layers.17.attention.o_proj.weight",
318
+ "vision_tower.transformer.layers.17.attention.q_proj.weight",
319
+ "vision_tower.transformer.layers.17.attention.v_proj.weight",
320
+ "vision_tower.transformer.layers.17.feed_forward.down_proj.weight",
321
+ "vision_tower.transformer.layers.17.feed_forward.gate_proj.weight",
322
+ "vision_tower.transformer.layers.17.feed_forward.up_proj.weight",
323
+ "vision_tower.transformer.layers.18.attention.k_proj.weight",
324
+ "vision_tower.transformer.layers.18.attention.o_proj.weight",
325
+ "vision_tower.transformer.layers.18.attention.q_proj.weight",
326
+ "vision_tower.transformer.layers.18.attention.v_proj.weight",
327
+ "vision_tower.transformer.layers.18.feed_forward.down_proj.weight",
328
+ "vision_tower.transformer.layers.18.feed_forward.gate_proj.weight",
329
+ "vision_tower.transformer.layers.18.feed_forward.up_proj.weight",
330
+ "vision_tower.transformer.layers.19.attention.k_proj.weight",
331
+ "vision_tower.transformer.layers.19.attention.o_proj.weight",
332
+ "vision_tower.transformer.layers.19.attention.q_proj.weight",
333
+ "vision_tower.transformer.layers.19.attention.v_proj.weight",
334
+ "vision_tower.transformer.layers.19.feed_forward.down_proj.weight",
335
+ "vision_tower.transformer.layers.19.feed_forward.gate_proj.weight",
336
+ "vision_tower.transformer.layers.19.feed_forward.up_proj.weight",
337
+ "vision_tower.transformer.layers.20.attention.k_proj.weight",
338
+ "vision_tower.transformer.layers.20.attention.o_proj.weight",
339
+ "vision_tower.transformer.layers.20.attention.q_proj.weight",
340
+ "vision_tower.transformer.layers.20.attention.v_proj.weight",
341
+ "vision_tower.transformer.layers.20.feed_forward.down_proj.weight",
342
+ "vision_tower.transformer.layers.20.feed_forward.gate_proj.weight",
343
+ "vision_tower.transformer.layers.20.feed_forward.up_proj.weight",
344
+ "vision_tower.transformer.layers.21.attention.k_proj.weight",
345
+ "vision_tower.transformer.layers.21.attention.o_proj.weight",
346
+ "vision_tower.transformer.layers.21.attention.q_proj.weight",
347
+ "vision_tower.transformer.layers.21.attention.v_proj.weight",
348
+ "vision_tower.transformer.layers.21.feed_forward.down_proj.weight",
349
+ "vision_tower.transformer.layers.21.feed_forward.gate_proj.weight",
350
+ "vision_tower.transformer.layers.21.feed_forward.up_proj.weight",
351
+ "vision_tower.transformer.layers.22.attention.k_proj.weight",
352
+ "vision_tower.transformer.layers.22.attention.o_proj.weight",
353
+ "vision_tower.transformer.layers.22.attention.q_proj.weight",
354
+ "vision_tower.transformer.layers.22.attention.v_proj.weight",
355
+ "vision_tower.transformer.layers.22.feed_forward.down_proj.weight",
356
+ "vision_tower.transformer.layers.22.feed_forward.gate_proj.weight",
357
+ "vision_tower.transformer.layers.22.feed_forward.up_proj.weight",
358
+ "vision_tower.transformer.layers.23.attention.k_proj.weight",
359
+ "vision_tower.transformer.layers.23.attention.o_proj.weight",
360
+ "vision_tower.transformer.layers.23.attention.q_proj.weight",
361
+ "vision_tower.transformer.layers.23.attention.v_proj.weight",
362
+ "vision_tower.transformer.layers.23.feed_forward.gate_proj.weight",
363
+ "vision_tower.transformer.layers.23.feed_forward.up_proj.weight"
364
+ ],
365
+ "uint8": [
366
+ "language_model.layers.0.self_attn.o_proj.weight",
367
+ "vision_tower.transformer.layers.1.attention.v_proj.weight",
368
+ "vision_tower.transformer.layers.1.feed_forward.down_proj.weight",
369
+ "vision_tower.transformer.layers.2.feed_forward.down_proj.weight",
370
+ "vision_tower.transformer.layers.3.attention.k_proj.weight",
371
+ "vision_tower.transformer.layers.3.attention.o_proj.weight",
372
+ "vision_tower.transformer.layers.4.attention.k_proj.weight",
373
+ "vision_tower.transformer.layers.4.attention.o_proj.weight",
374
+ "vision_tower.transformer.layers.5.attention.k_proj.weight"
375
+ ]
376
+ },
377
+ "modules_quant_config": {},
378
+ "modules_to_not_convert": [
379
+ "lm_head",
380
+ "time_text_embed",
381
+ ".proj_out",
382
+ ".emb_out",
383
+ ".norm_out",
384
+ ".vid_in",
385
+ ".txt_in",
386
+ ".x_embedder",
387
+ ".time_embed",
388
+ ".img_out",
389
+ ".txt_out",
390
+ "patch_embed",
391
+ "patch_embedding",
392
+ ".emb_in",
393
+ ".t_embedder",
394
+ ".condition_embedder",
395
+ ".context_embedder",
396
+ ".vid_out",
397
+ "multi_modal_projector",
398
+ ".y_embedder",
399
+ "patch_emb",
400
+ "wte",
401
+ "language_model.norm",
402
+ ".img_in",
403
+ ".final_layer"
404
+ ],
405
+ "non_blocking": false,
406
+ "quant_conv": false,
407
+ "quant_method": "sdnq",
408
+ "quantization_device": null,
409
+ "quantized_matmul_dtype": null,
410
+ "return_device": null,
411
+ "sdnq_version": "0.1.6",
412
+ "svd_rank": 32,
413
+ "svd_steps": 8,
414
+ "use_dynamic_quantization": true,
415
+ "use_grad_ckpt": true,
416
+ "use_quantized_matmul": false,
417
+ "use_quantized_matmul_conv": false,
418
+ "use_static_quantization": true,
419
+ "use_stochastic_rounding": false,
420
+ "use_svd": false,
421
+ "weights_dtype": "int8"
422
+ },
423
+ "spatial_merge_size": 2,
424
+ "text_config": {
425
+ "attention_dropout": 0.0,
426
+ "bos_token_id": 1,
427
+ "dtype": "bfloat16",
428
+ "eos_token_id": 2,
429
+ "head_dim": 128,
430
+ "hidden_act": "silu",
431
+ "hidden_size": 3072,
432
+ "initializer_range": 0.02,
433
+ "intermediate_size": 9216,
434
+ "max_position_embeddings": 262144,
435
+ "model_type": "ministral3",
436
+ "num_attention_heads": 32,
437
+ "num_hidden_layers": 26,
438
+ "num_key_value_heads": 8,
439
+ "pad_token_id": 11,
440
+ "rms_norm_eps": 1e-05,
441
+ "rope_parameters": {
442
+ "beta_fast": 32.0,
443
+ "beta_slow": 1.0,
444
+ "factor": 16.0,
445
+ "llama_4_scaling_beta": 0.1,
446
+ "mscale": 1.0,
447
+ "mscale_all_dim": 1.0,
448
+ "original_max_position_embeddings": 16384,
449
+ "rope_theta": 1000000.0,
450
+ "rope_type": "yarn",
451
+ "type": "yarn"
452
+ },
453
+ "sliding_window": null,
454
+ "tie_word_embeddings": true,
455
+ "use_cache": true,
456
+ "vocab_size": 131072
457
+ },
458
+ "tie_word_embeddings": true,
459
+ "transformers_version": "5.3.0",
460
+ "vision_config": {
461
+ "attention_dropout": 0.0,
462
+ "dtype": "bfloat16",
463
+ "head_dim": 64,
464
+ "hidden_act": "silu",
465
+ "hidden_size": 1024,
466
+ "image_size": 1540,
467
+ "initializer_range": 0.02,
468
+ "intermediate_size": 4096,
469
+ "model_type": "pixtral",
470
+ "num_attention_heads": 16,
471
+ "num_channels": 3,
472
+ "num_hidden_layers": 24,
473
+ "patch_size": 14,
474
+ "rope_parameters": {
475
+ "rope_theta": 10000.0,
476
+ "rope_type": "default"
477
+ }
478
+ },
479
+ "vision_feature_layer": -1
480
+ }
text_encoder/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e896546ba39bb1878a3b315620d0867cc991247842186d231c5c197993fe90c
3
+ size 4279597752
tokenizer/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:577575622324b2e099e2648be26bdeb5e5815ffe66d7004e9e3ddbf421db6bf1
3
+ size 17078110
tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": null,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "</s>",
7
+ "is_local": true,
8
+ "legacy": true,
9
+ "model_max_length": 2048,
10
+ "pad_token": "<pad>",
11
+ "processor_class": "PixtralProcessor",
12
+ "tokenizer_class": "TokenizersBackend",
13
+ "unk_token": "<unk>",
14
+ "use_default_system_prompt": false
15
+ }
transformer/config.json ADDED
@@ -0,0 +1,339 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "ErnieImageTransformer2DModel",
3
+ "_diffusers_version": "0.38.0.dev0",
4
+ "_name_or_path": "Baidu/ERNIE-Image",
5
+ "eps": 1e-06,
6
+ "ffn_hidden_size": 12288,
7
+ "hidden_size": 4096,
8
+ "in_channels": 128,
9
+ "num_attention_heads": 32,
10
+ "num_layers": 36,
11
+ "out_channels": 128,
12
+ "patch_size": 1,
13
+ "qk_layernorm": true,
14
+ "quantization_config": {
15
+ "add_skip_keys": true,
16
+ "dequantize_fp32": false,
17
+ "dynamic_loss_threshold": 0.0001,
18
+ "group_size": 0,
19
+ "is_integer": true,
20
+ "is_training": false,
21
+ "modules_dtype_dict": {
22
+ "float10_e4m5fn": [
23
+ "final_linear.weight",
24
+ "final_norm.linear.weight"
25
+ ],
26
+ "float9_e3m5fn": [
27
+ "adaLN_modulation.1.weight",
28
+ "layers.24.self_attention.to_k.weight",
29
+ "layers.29.self_attention.to_k.weight",
30
+ "layers.30.self_attention.to_k.weight",
31
+ "layers.31.self_attention.to_k.weight",
32
+ "layers.33.self_attention.to_k.weight",
33
+ "layers.34.self_attention.to_k.weight",
34
+ "layers.35.mlp.linear_fc2.weight",
35
+ "layers.35.self_attention.to_k.weight"
36
+ ],
37
+ "int8": [
38
+ "layers.0.mlp.gate_proj.weight",
39
+ "layers.0.mlp.linear_fc2.weight",
40
+ "layers.0.mlp.up_proj.weight",
41
+ "layers.0.self_attention.to_k.weight",
42
+ "layers.0.self_attention.to_out.0.weight",
43
+ "layers.0.self_attention.to_q.weight",
44
+ "layers.0.self_attention.to_v.weight",
45
+ "layers.1.mlp.gate_proj.weight",
46
+ "layers.1.mlp.linear_fc2.weight",
47
+ "layers.1.mlp.up_proj.weight",
48
+ "layers.1.self_attention.to_k.weight",
49
+ "layers.1.self_attention.to_out.0.weight",
50
+ "layers.1.self_attention.to_q.weight",
51
+ "layers.1.self_attention.to_v.weight",
52
+ "layers.10.mlp.gate_proj.weight",
53
+ "layers.10.mlp.linear_fc2.weight",
54
+ "layers.10.mlp.up_proj.weight",
55
+ "layers.10.self_attention.to_k.weight",
56
+ "layers.10.self_attention.to_out.0.weight",
57
+ "layers.10.self_attention.to_q.weight",
58
+ "layers.10.self_attention.to_v.weight",
59
+ "layers.11.mlp.gate_proj.weight",
60
+ "layers.11.mlp.linear_fc2.weight",
61
+ "layers.11.mlp.up_proj.weight",
62
+ "layers.11.self_attention.to_k.weight",
63
+ "layers.11.self_attention.to_out.0.weight",
64
+ "layers.11.self_attention.to_q.weight",
65
+ "layers.11.self_attention.to_v.weight",
66
+ "layers.12.mlp.gate_proj.weight",
67
+ "layers.12.mlp.linear_fc2.weight",
68
+ "layers.12.mlp.up_proj.weight",
69
+ "layers.12.self_attention.to_k.weight",
70
+ "layers.12.self_attention.to_out.0.weight",
71
+ "layers.12.self_attention.to_q.weight",
72
+ "layers.12.self_attention.to_v.weight",
73
+ "layers.13.mlp.gate_proj.weight",
74
+ "layers.13.mlp.linear_fc2.weight",
75
+ "layers.13.mlp.up_proj.weight",
76
+ "layers.13.self_attention.to_k.weight",
77
+ "layers.13.self_attention.to_out.0.weight",
78
+ "layers.13.self_attention.to_q.weight",
79
+ "layers.13.self_attention.to_v.weight",
80
+ "layers.14.mlp.gate_proj.weight",
81
+ "layers.14.mlp.linear_fc2.weight",
82
+ "layers.14.mlp.up_proj.weight",
83
+ "layers.14.self_attention.to_k.weight",
84
+ "layers.14.self_attention.to_out.0.weight",
85
+ "layers.14.self_attention.to_q.weight",
86
+ "layers.14.self_attention.to_v.weight",
87
+ "layers.15.mlp.gate_proj.weight",
88
+ "layers.15.mlp.linear_fc2.weight",
89
+ "layers.15.mlp.up_proj.weight",
90
+ "layers.15.self_attention.to_k.weight",
91
+ "layers.15.self_attention.to_out.0.weight",
92
+ "layers.15.self_attention.to_q.weight",
93
+ "layers.15.self_attention.to_v.weight",
94
+ "layers.16.mlp.gate_proj.weight",
95
+ "layers.16.mlp.linear_fc2.weight",
96
+ "layers.16.mlp.up_proj.weight",
97
+ "layers.16.self_attention.to_k.weight",
98
+ "layers.16.self_attention.to_out.0.weight",
99
+ "layers.16.self_attention.to_q.weight",
100
+ "layers.16.self_attention.to_v.weight",
101
+ "layers.17.mlp.gate_proj.weight",
102
+ "layers.17.mlp.linear_fc2.weight",
103
+ "layers.17.mlp.up_proj.weight",
104
+ "layers.17.self_attention.to_k.weight",
105
+ "layers.17.self_attention.to_out.0.weight",
106
+ "layers.17.self_attention.to_q.weight",
107
+ "layers.17.self_attention.to_v.weight",
108
+ "layers.18.mlp.gate_proj.weight",
109
+ "layers.18.mlp.linear_fc2.weight",
110
+ "layers.18.mlp.up_proj.weight",
111
+ "layers.18.self_attention.to_k.weight",
112
+ "layers.18.self_attention.to_out.0.weight",
113
+ "layers.18.self_attention.to_q.weight",
114
+ "layers.18.self_attention.to_v.weight",
115
+ "layers.19.mlp.gate_proj.weight",
116
+ "layers.19.mlp.linear_fc2.weight",
117
+ "layers.19.mlp.up_proj.weight",
118
+ "layers.19.self_attention.to_out.0.weight",
119
+ "layers.19.self_attention.to_q.weight",
120
+ "layers.19.self_attention.to_v.weight",
121
+ "layers.2.mlp.gate_proj.weight",
122
+ "layers.2.mlp.linear_fc2.weight",
123
+ "layers.2.mlp.up_proj.weight",
124
+ "layers.2.self_attention.to_k.weight",
125
+ "layers.2.self_attention.to_out.0.weight",
126
+ "layers.2.self_attention.to_q.weight",
127
+ "layers.2.self_attention.to_v.weight",
128
+ "layers.20.mlp.gate_proj.weight",
129
+ "layers.20.mlp.linear_fc2.weight",
130
+ "layers.20.mlp.up_proj.weight",
131
+ "layers.20.self_attention.to_out.0.weight",
132
+ "layers.20.self_attention.to_v.weight",
133
+ "layers.21.mlp.gate_proj.weight",
134
+ "layers.21.mlp.linear_fc2.weight",
135
+ "layers.21.mlp.up_proj.weight",
136
+ "layers.21.self_attention.to_out.0.weight",
137
+ "layers.21.self_attention.to_q.weight",
138
+ "layers.21.self_attention.to_v.weight",
139
+ "layers.22.self_attention.to_q.weight",
140
+ "layers.22.self_attention.to_v.weight",
141
+ "layers.3.mlp.gate_proj.weight",
142
+ "layers.3.mlp.linear_fc2.weight",
143
+ "layers.3.mlp.up_proj.weight",
144
+ "layers.3.self_attention.to_k.weight",
145
+ "layers.3.self_attention.to_out.0.weight",
146
+ "layers.3.self_attention.to_q.weight",
147
+ "layers.3.self_attention.to_v.weight",
148
+ "layers.4.mlp.gate_proj.weight",
149
+ "layers.4.mlp.linear_fc2.weight",
150
+ "layers.4.mlp.up_proj.weight",
151
+ "layers.4.self_attention.to_k.weight",
152
+ "layers.4.self_attention.to_out.0.weight",
153
+ "layers.4.self_attention.to_q.weight",
154
+ "layers.4.self_attention.to_v.weight",
155
+ "layers.5.mlp.gate_proj.weight",
156
+ "layers.5.mlp.linear_fc2.weight",
157
+ "layers.5.mlp.up_proj.weight",
158
+ "layers.5.self_attention.to_k.weight",
159
+ "layers.5.self_attention.to_out.0.weight",
160
+ "layers.5.self_attention.to_q.weight",
161
+ "layers.5.self_attention.to_v.weight",
162
+ "layers.6.mlp.gate_proj.weight",
163
+ "layers.6.mlp.linear_fc2.weight",
164
+ "layers.6.mlp.up_proj.weight",
165
+ "layers.6.self_attention.to_k.weight",
166
+ "layers.6.self_attention.to_out.0.weight",
167
+ "layers.6.self_attention.to_q.weight",
168
+ "layers.6.self_attention.to_v.weight",
169
+ "layers.7.mlp.gate_proj.weight",
170
+ "layers.7.mlp.up_proj.weight",
171
+ "layers.7.self_attention.to_k.weight",
172
+ "layers.7.self_attention.to_out.0.weight",
173
+ "layers.7.self_attention.to_q.weight",
174
+ "layers.7.self_attention.to_v.weight",
175
+ "layers.8.mlp.gate_proj.weight",
176
+ "layers.8.mlp.linear_fc2.weight",
177
+ "layers.8.mlp.up_proj.weight",
178
+ "layers.8.self_attention.to_k.weight",
179
+ "layers.8.self_attention.to_out.0.weight",
180
+ "layers.8.self_attention.to_q.weight",
181
+ "layers.8.self_attention.to_v.weight",
182
+ "layers.9.mlp.gate_proj.weight",
183
+ "layers.9.mlp.linear_fc2.weight",
184
+ "layers.9.mlp.up_proj.weight",
185
+ "layers.9.self_attention.to_k.weight",
186
+ "layers.9.self_attention.to_out.0.weight",
187
+ "layers.9.self_attention.to_q.weight",
188
+ "layers.9.self_attention.to_v.weight",
189
+ "text_proj.weight",
190
+ "layers.22.mlp.gate_proj.weight",
191
+ "layers.22.mlp.linear_fc2.weight",
192
+ "layers.22.mlp.up_proj.weight",
193
+ "layers.22.self_attention.to_out.0.weight",
194
+ "layers.23.mlp.gate_proj.weight",
195
+ "layers.23.mlp.linear_fc2.weight",
196
+ "layers.23.mlp.up_proj.weight",
197
+ "layers.23.self_attention.to_out.0.weight",
198
+ "layers.23.self_attention.to_q.weight",
199
+ "layers.23.self_attention.to_v.weight",
200
+ "layers.24.mlp.gate_proj.weight",
201
+ "layers.24.mlp.linear_fc2.weight",
202
+ "layers.24.mlp.up_proj.weight",
203
+ "layers.24.self_attention.to_out.0.weight",
204
+ "layers.24.self_attention.to_q.weight",
205
+ "layers.25.mlp.gate_proj.weight",
206
+ "layers.25.mlp.linear_fc2.weight",
207
+ "layers.25.mlp.up_proj.weight",
208
+ "layers.25.self_attention.to_out.0.weight",
209
+ "layers.25.self_attention.to_q.weight",
210
+ "layers.25.self_attention.to_v.weight",
211
+ "layers.26.mlp.gate_proj.weight",
212
+ "layers.26.mlp.linear_fc2.weight",
213
+ "layers.26.mlp.up_proj.weight",
214
+ "layers.26.self_attention.to_out.0.weight",
215
+ "layers.26.self_attention.to_v.weight",
216
+ "layers.27.mlp.gate_proj.weight",
217
+ "layers.27.mlp.linear_fc2.weight",
218
+ "layers.27.mlp.up_proj.weight",
219
+ "layers.27.self_attention.to_out.0.weight",
220
+ "layers.27.self_attention.to_v.weight",
221
+ "layers.28.mlp.gate_proj.weight",
222
+ "layers.28.mlp.linear_fc2.weight",
223
+ "layers.28.mlp.up_proj.weight",
224
+ "layers.28.self_attention.to_out.0.weight",
225
+ "layers.28.self_attention.to_v.weight",
226
+ "layers.29.mlp.gate_proj.weight",
227
+ "layers.29.mlp.linear_fc2.weight",
228
+ "layers.29.mlp.up_proj.weight",
229
+ "layers.29.self_attention.to_out.0.weight",
230
+ "layers.29.self_attention.to_v.weight",
231
+ "layers.30.mlp.linear_fc2.weight",
232
+ "layers.30.mlp.up_proj.weight",
233
+ "layers.30.self_attention.to_out.0.weight",
234
+ "layers.30.self_attention.to_v.weight",
235
+ "layers.31.mlp.linear_fc2.weight",
236
+ "layers.31.mlp.up_proj.weight",
237
+ "layers.31.self_attention.to_out.0.weight",
238
+ "layers.32.mlp.linear_fc2.weight",
239
+ "layers.32.mlp.up_proj.weight",
240
+ "layers.32.self_attention.to_out.0.weight",
241
+ "layers.32.self_attention.to_v.weight",
242
+ "layers.33.mlp.linear_fc2.weight",
243
+ "layers.33.mlp.up_proj.weight",
244
+ "layers.33.self_attention.to_out.0.weight",
245
+ "layers.33.self_attention.to_v.weight",
246
+ "layers.34.mlp.gate_proj.weight",
247
+ "layers.34.mlp.linear_fc2.weight",
248
+ "layers.34.mlp.up_proj.weight",
249
+ "layers.34.self_attention.to_out.0.weight",
250
+ "layers.35.mlp.up_proj.weight",
251
+ "layers.35.self_attention.to_out.0.weight"
252
+ ],
253
+ "uint8": [
254
+ "layers.19.self_attention.to_k.weight",
255
+ "layers.20.self_attention.to_k.weight",
256
+ "layers.20.self_attention.to_q.weight",
257
+ "layers.21.self_attention.to_k.weight",
258
+ "layers.22.self_attention.to_k.weight",
259
+ "layers.7.mlp.linear_fc2.weight",
260
+ "layers.23.self_attention.to_k.weight",
261
+ "layers.24.self_attention.to_v.weight",
262
+ "layers.25.self_attention.to_k.weight",
263
+ "layers.26.self_attention.to_k.weight",
264
+ "layers.26.self_attention.to_q.weight",
265
+ "layers.27.self_attention.to_k.weight",
266
+ "layers.27.self_attention.to_q.weight",
267
+ "layers.28.self_attention.to_k.weight",
268
+ "layers.28.self_attention.to_q.weight",
269
+ "layers.29.self_attention.to_q.weight",
270
+ "layers.30.mlp.gate_proj.weight",
271
+ "layers.30.self_attention.to_q.weight",
272
+ "layers.31.mlp.gate_proj.weight",
273
+ "layers.31.self_attention.to_q.weight",
274
+ "layers.31.self_attention.to_v.weight",
275
+ "layers.32.mlp.gate_proj.weight",
276
+ "layers.32.self_attention.to_k.weight",
277
+ "layers.32.self_attention.to_q.weight",
278
+ "layers.33.mlp.gate_proj.weight",
279
+ "layers.33.self_attention.to_q.weight",
280
+ "layers.34.self_attention.to_q.weight",
281
+ "layers.34.self_attention.to_v.weight",
282
+ "layers.35.mlp.gate_proj.weight",
283
+ "layers.35.self_attention.to_q.weight",
284
+ "layers.35.self_attention.to_v.weight"
285
+ ]
286
+ },
287
+ "modules_quant_config": {},
288
+ "modules_to_not_convert": [
289
+ "lm_head",
290
+ "time_text_embed",
291
+ ".proj_out",
292
+ ".emb_out",
293
+ ".norm_out",
294
+ ".vid_in",
295
+ ".txt_in",
296
+ ".x_embedder",
297
+ ".time_embed",
298
+ ".img_out",
299
+ ".txt_out",
300
+ "patch_embed",
301
+ "patch_embedding",
302
+ ".emb_in",
303
+ ".t_embedder",
304
+ ".condition_embedder",
305
+ ".context_embedder",
306
+ ".vid_out",
307
+ "multi_modal_projector",
308
+ ".y_embedder",
309
+ "patch_emb",
310
+ "wte",
311
+ ".img_in",
312
+ ".final_layer"
313
+ ],
314
+ "non_blocking": false,
315
+ "quant_conv": false,
316
+ "quant_method": "sdnq",
317
+ "quantization_device": null,
318
+ "quantized_matmul_dtype": null,
319
+ "return_device": null,
320
+ "sdnq_version": "0.1.6",
321
+ "svd_rank": 32,
322
+ "svd_steps": 8,
323
+ "use_dynamic_quantization": true,
324
+ "use_grad_ckpt": true,
325
+ "use_quantized_matmul": false,
326
+ "use_quantized_matmul_conv": false,
327
+ "use_static_quantization": true,
328
+ "use_stochastic_rounding": false,
329
+ "use_svd": false,
330
+ "weights_dtype": "int8"
331
+ },
332
+ "rope_axes_dim": [
333
+ 32,
334
+ 48,
335
+ 48
336
+ ],
337
+ "rope_theta": 256,
338
+ "text_in_dim": 3072
339
+ }
transformer/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:414fd7f2c6fab41f7cb0d78f0e1fdf25a4e4e2a06c901a0da249f2baf35b772d
3
+ size 8126787072
vae/config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "AutoencoderKLFlux2",
3
+ "_diffusers_version": "0.36.0",
4
+ "_name_or_path": "./weights/black-forest-labs/FLUX.2-dev/vae/",
5
+ "act_fn": "silu",
6
+ "batch_norm_eps": 0.0001,
7
+ "batch_norm_momentum": 0.1,
8
+ "block_out_channels": [
9
+ 128,
10
+ 256,
11
+ 512,
12
+ 512
13
+ ],
14
+ "down_block_types": [
15
+ "DownEncoderBlock2D",
16
+ "DownEncoderBlock2D",
17
+ "DownEncoderBlock2D",
18
+ "DownEncoderBlock2D"
19
+ ],
20
+ "force_upcast": true,
21
+ "in_channels": 3,
22
+ "latent_channels": 32,
23
+ "layers_per_block": 2,
24
+ "mid_block_add_attention": true,
25
+ "norm_num_groups": 32,
26
+ "out_channels": 3,
27
+ "patch_size": [
28
+ 2,
29
+ 2
30
+ ],
31
+ "sample_size": 1024,
32
+ "up_block_types": [
33
+ "UpDecoderBlock2D",
34
+ "UpDecoderBlock2D",
35
+ "UpDecoderBlock2D",
36
+ "UpDecoderBlock2D"
37
+ ],
38
+ "use_post_quant_conv": true,
39
+ "use_quant_conv": true
40
+ }
vae/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca70d2202afe6415bdbcb8793ba8cd99fd159cfe6192381504d6c4d3036e0f04
3
+ size 168120878