Upload folder using huggingface_hub
Browse files- config.json +190 -4
- model-00001-of-00007.safetensors +2 -2
- model-00002-of-00007.safetensors +2 -2
- model-00003-of-00007.safetensors +2 -2
- model-00004-of-00007.safetensors +2 -2
- model-00005-of-00007.safetensors +2 -2
- model-00006-of-00007.safetensors +2 -2
- model-00007-of-00007.safetensors +2 -2
- model.safetensors.index.json +0 -0
- preprocessor_config.json +26 -0
- processor_config.json +6 -0
- tokenizer_config.json +2 -0
config.json
CHANGED
|
@@ -1,4 +1,5 @@
|
|
| 1 |
{
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"KimiVLForConditionalGeneration"
|
| 4 |
],
|
|
@@ -7,9 +8,56 @@
|
|
| 7 |
"AutoModel": "modeling_kimi_vl.KimiVLForConditionalGeneration",
|
| 8 |
"AutoModelForCausalLM": "modeling_kimi_vl.KimiVLForConditionalGeneration"
|
| 9 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
"ignore_index": -100,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
"media_placeholder_token_id": 163605,
|
|
|
|
| 12 |
"model_type": "kimi_vl",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
"text_config": {
|
| 14 |
"vocab_size": 163840,
|
| 15 |
"max_position_embeddings": 131072,
|
|
@@ -17,6 +65,7 @@
|
|
| 17 |
"intermediate_size": 11264,
|
| 18 |
"moe_intermediate_size": 1408,
|
| 19 |
"num_hidden_layers": 27,
|
|
|
|
| 20 |
"num_attention_heads": 16,
|
| 21 |
"n_shared_experts": 2,
|
| 22 |
"n_routed_experts": 64,
|
|
@@ -47,14 +96,151 @@
|
|
| 47 |
"rope_scaling": null,
|
| 48 |
"attention_bias": false,
|
| 49 |
"attention_dropout": 0.0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
"bos_token_id": 163584,
|
| 51 |
"pad_token_id": 163839,
|
| 52 |
"eos_token_id": 163585,
|
| 53 |
-
"
|
| 54 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
},
|
|
|
|
|
|
|
| 56 |
"tie_word_embeddings": false,
|
| 57 |
-
"
|
| 58 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
"vocab_size": 163840
|
| 60 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
"architectures": [
|
| 4 |
"KimiVLForConditionalGeneration"
|
| 5 |
],
|
|
|
|
| 8 |
"AutoModel": "modeling_kimi_vl.KimiVLForConditionalGeneration",
|
| 9 |
"AutoModelForCausalLM": "modeling_kimi_vl.KimiVLForConditionalGeneration"
|
| 10 |
},
|
| 11 |
+
"bad_words_ids": null,
|
| 12 |
+
"begin_suppress_tokens": null,
|
| 13 |
+
"bos_token_id": null,
|
| 14 |
+
"chunk_size_feed_forward": 0,
|
| 15 |
+
"cross_attention_hidden_size": null,
|
| 16 |
+
"decoder_start_token_id": null,
|
| 17 |
+
"diversity_penalty": 0.0,
|
| 18 |
+
"do_sample": false,
|
| 19 |
+
"early_stopping": false,
|
| 20 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 21 |
+
"eos_token_id": null,
|
| 22 |
+
"exponential_decay_length_penalty": null,
|
| 23 |
+
"finetuning_task": null,
|
| 24 |
+
"forced_bos_token_id": null,
|
| 25 |
+
"forced_eos_token_id": null,
|
| 26 |
+
"id2label": {
|
| 27 |
+
"0": "LABEL_0",
|
| 28 |
+
"1": "LABEL_1"
|
| 29 |
+
},
|
| 30 |
"ignore_index": -100,
|
| 31 |
+
"is_decoder": false,
|
| 32 |
+
"is_encoder_decoder": false,
|
| 33 |
+
"label2id": {
|
| 34 |
+
"LABEL_0": 0,
|
| 35 |
+
"LABEL_1": 1
|
| 36 |
+
},
|
| 37 |
+
"length_penalty": 1.0,
|
| 38 |
+
"max_length": 20,
|
| 39 |
"media_placeholder_token_id": 163605,
|
| 40 |
+
"min_length": 0,
|
| 41 |
"model_type": "kimi_vl",
|
| 42 |
+
"no_repeat_ngram_size": 0,
|
| 43 |
+
"num_beam_groups": 1,
|
| 44 |
+
"num_beams": 1,
|
| 45 |
+
"num_return_sequences": 1,
|
| 46 |
+
"output_attentions": false,
|
| 47 |
+
"output_hidden_states": false,
|
| 48 |
+
"output_scores": false,
|
| 49 |
+
"pad_token_id": 0,
|
| 50 |
+
"prefix": null,
|
| 51 |
+
"problem_type": null,
|
| 52 |
+
"pruned_heads": {},
|
| 53 |
+
"remove_invalid_values": false,
|
| 54 |
+
"repetition_penalty": 1.0,
|
| 55 |
+
"return_dict": true,
|
| 56 |
+
"return_dict_in_generate": false,
|
| 57 |
+
"sep_token_id": null,
|
| 58 |
+
"suppress_tokens": null,
|
| 59 |
+
"task_specific_params": null,
|
| 60 |
+
"temperature": 1.0,
|
| 61 |
"text_config": {
|
| 62 |
"vocab_size": 163840,
|
| 63 |
"max_position_embeddings": 131072,
|
|
|
|
| 65 |
"intermediate_size": 11264,
|
| 66 |
"moe_intermediate_size": 1408,
|
| 67 |
"num_hidden_layers": 27,
|
| 68 |
+
"num_nextn_predict_layers": 1,
|
| 69 |
"num_attention_heads": 16,
|
| 70 |
"n_shared_experts": 2,
|
| 71 |
"n_routed_experts": 64,
|
|
|
|
| 96 |
"rope_scaling": null,
|
| 97 |
"attention_bias": false,
|
| 98 |
"attention_dropout": 0.0,
|
| 99 |
+
"return_dict": true,
|
| 100 |
+
"output_hidden_states": false,
|
| 101 |
+
"torchscript": false,
|
| 102 |
+
"torch_dtype": "bfloat16",
|
| 103 |
+
"use_bfloat16": false,
|
| 104 |
+
"tf_legacy_loss": false,
|
| 105 |
+
"pruned_heads": {},
|
| 106 |
+
"tie_word_embeddings": false,
|
| 107 |
+
"chunk_size_feed_forward": 0,
|
| 108 |
+
"is_encoder_decoder": false,
|
| 109 |
+
"is_decoder": false,
|
| 110 |
+
"cross_attention_hidden_size": null,
|
| 111 |
+
"add_cross_attention": false,
|
| 112 |
+
"tie_encoder_decoder": false,
|
| 113 |
+
"max_length": 20,
|
| 114 |
+
"min_length": 0,
|
| 115 |
+
"do_sample": false,
|
| 116 |
+
"early_stopping": false,
|
| 117 |
+
"num_beams": 1,
|
| 118 |
+
"num_beam_groups": 1,
|
| 119 |
+
"diversity_penalty": 0.0,
|
| 120 |
+
"temperature": 1.0,
|
| 121 |
+
"top_k": 50,
|
| 122 |
+
"top_p": 1.0,
|
| 123 |
+
"typical_p": 1.0,
|
| 124 |
+
"repetition_penalty": 1.0,
|
| 125 |
+
"length_penalty": 1.0,
|
| 126 |
+
"no_repeat_ngram_size": 0,
|
| 127 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 128 |
+
"bad_words_ids": null,
|
| 129 |
+
"num_return_sequences": 1,
|
| 130 |
+
"output_scores": false,
|
| 131 |
+
"return_dict_in_generate": false,
|
| 132 |
+
"forced_bos_token_id": null,
|
| 133 |
+
"forced_eos_token_id": null,
|
| 134 |
+
"remove_invalid_values": false,
|
| 135 |
+
"exponential_decay_length_penalty": null,
|
| 136 |
+
"suppress_tokens": null,
|
| 137 |
+
"begin_suppress_tokens": null,
|
| 138 |
+
"architectures": null,
|
| 139 |
+
"finetuning_task": null,
|
| 140 |
+
"id2label": {
|
| 141 |
+
"0": "LABEL_0",
|
| 142 |
+
"1": "LABEL_1"
|
| 143 |
+
},
|
| 144 |
+
"label2id": {
|
| 145 |
+
"LABEL_0": 0,
|
| 146 |
+
"LABEL_1": 1
|
| 147 |
+
},
|
| 148 |
+
"tokenizer_class": null,
|
| 149 |
+
"prefix": null,
|
| 150 |
"bos_token_id": 163584,
|
| 151 |
"pad_token_id": 163839,
|
| 152 |
"eos_token_id": 163585,
|
| 153 |
+
"sep_token_id": null,
|
| 154 |
+
"decoder_start_token_id": null,
|
| 155 |
+
"task_specific_params": null,
|
| 156 |
+
"problem_type": null,
|
| 157 |
+
"_name_or_path": "",
|
| 158 |
+
"model_type": "deepseek_v3",
|
| 159 |
+
"output_attentions": false
|
| 160 |
},
|
| 161 |
+
"tf_legacy_loss": false,
|
| 162 |
+
"tie_encoder_decoder": false,
|
| 163 |
"tie_word_embeddings": false,
|
| 164 |
+
"tokenizer_class": null,
|
| 165 |
+
"top_k": 50,
|
| 166 |
+
"top_p": 1.0,
|
| 167 |
+
"torchscript": false,
|
| 168 |
+
"transformers_version": "4.53.3",
|
| 169 |
+
"typical_p": 1.0,
|
| 170 |
+
"use_bfloat16": false,
|
| 171 |
+
"vision_config": {
|
| 172 |
+
"return_dict": true,
|
| 173 |
+
"output_hidden_states": false,
|
| 174 |
+
"torchscript": false,
|
| 175 |
+
"torch_dtype": "bfloat16",
|
| 176 |
+
"use_bfloat16": false,
|
| 177 |
+
"tf_legacy_loss": false,
|
| 178 |
+
"pruned_heads": {},
|
| 179 |
+
"tie_word_embeddings": true,
|
| 180 |
+
"chunk_size_feed_forward": 0,
|
| 181 |
+
"is_encoder_decoder": false,
|
| 182 |
+
"is_decoder": false,
|
| 183 |
+
"cross_attention_hidden_size": null,
|
| 184 |
+
"add_cross_attention": false,
|
| 185 |
+
"tie_encoder_decoder": false,
|
| 186 |
+
"max_length": 20,
|
| 187 |
+
"min_length": 0,
|
| 188 |
+
"do_sample": false,
|
| 189 |
+
"early_stopping": false,
|
| 190 |
+
"num_beams": 1,
|
| 191 |
+
"num_beam_groups": 1,
|
| 192 |
+
"diversity_penalty": 0.0,
|
| 193 |
+
"temperature": 1.0,
|
| 194 |
+
"top_k": 50,
|
| 195 |
+
"top_p": 1.0,
|
| 196 |
+
"typical_p": 1.0,
|
| 197 |
+
"repetition_penalty": 1.0,
|
| 198 |
+
"length_penalty": 1.0,
|
| 199 |
+
"no_repeat_ngram_size": 0,
|
| 200 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 201 |
+
"bad_words_ids": null,
|
| 202 |
+
"num_return_sequences": 1,
|
| 203 |
+
"output_scores": false,
|
| 204 |
+
"return_dict_in_generate": false,
|
| 205 |
+
"forced_bos_token_id": null,
|
| 206 |
+
"forced_eos_token_id": null,
|
| 207 |
+
"remove_invalid_values": false,
|
| 208 |
+
"exponential_decay_length_penalty": null,
|
| 209 |
+
"suppress_tokens": null,
|
| 210 |
+
"begin_suppress_tokens": null,
|
| 211 |
+
"architectures": null,
|
| 212 |
+
"finetuning_task": null,
|
| 213 |
+
"id2label": {
|
| 214 |
+
"0": "LABEL_0",
|
| 215 |
+
"1": "LABEL_1"
|
| 216 |
+
},
|
| 217 |
+
"label2id": {
|
| 218 |
+
"LABEL_0": 0,
|
| 219 |
+
"LABEL_1": 1
|
| 220 |
+
},
|
| 221 |
+
"tokenizer_class": null,
|
| 222 |
+
"prefix": null,
|
| 223 |
+
"bos_token_id": null,
|
| 224 |
+
"pad_token_id": null,
|
| 225 |
+
"eos_token_id": null,
|
| 226 |
+
"sep_token_id": null,
|
| 227 |
+
"decoder_start_token_id": null,
|
| 228 |
+
"task_specific_params": null,
|
| 229 |
+
"problem_type": null,
|
| 230 |
+
"_name_or_path": "",
|
| 231 |
+
"model_type": "moonvit",
|
| 232 |
+
"patch_size": 14,
|
| 233 |
+
"init_pos_emb_height": 64,
|
| 234 |
+
"init_pos_emb_width": 64,
|
| 235 |
+
"num_hidden_layers": 27,
|
| 236 |
+
"num_attention_heads": 16,
|
| 237 |
+
"hidden_size": 1152,
|
| 238 |
+
"intermediate_size": 4304,
|
| 239 |
+
"merge_kernel_size": [
|
| 240 |
+
2,
|
| 241 |
+
2
|
| 242 |
+
],
|
| 243 |
+
"output_attentions": false
|
| 244 |
+
},
|
| 245 |
"vocab_size": 163840
|
| 246 |
}
|
model-00001-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c2c40ab7ee663d1a9d8dfdf886f58fc1ed72360d3d040134903ab4a7d4f82c7f
|
| 3 |
+
size 5207425600
|
model-00002-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99400fb4a4db93191ee2f0eea827f5dbde491460729e221adcefef959d7934e3
|
| 3 |
+
size 5047890573
|
model-00003-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0b38aa086aa1bc5ec73f5a12065433f8e21a1cff8cb6586c1946d46b7b8e35f
|
| 3 |
+
size 5047890606
|
model-00004-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0a8d9c435f8715067d9cf9b359a4226d5fa17e2a4c55c502d856a1cd8587511
|
| 3 |
+
size 5110291870
|
model-00005-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eca1fc288e7134dd07f4d15e9f2c457aa1fed551e39db8b60c2bc752cd53b98a
|
| 3 |
+
size 5047890618
|
model-00006-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d1452ab13cc898a14c88ffc00a1e7226372d349707493dcffe58a7b718287de
|
| 3 |
+
size 5047890650
|
model-00007-of-00007.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:625d4b768aa82c2e33050f24ea937de6def4937c1040ca60f35e0258f8de2ad9
|
| 3 |
+
size 2306124504
|
model.safetensors.index.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_map": {
|
| 3 |
+
"AutoImageProcessor": "image_processing_kimi_vl.KimiVLImageProcessor",
|
| 4 |
+
"AutoProcessor": "processing_kimi_vl.KimiVLProcessor"
|
| 5 |
+
},
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "KimiVLImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"in_token_limit": 16384,
|
| 18 |
+
"merge_kernel_size": [
|
| 19 |
+
2,
|
| 20 |
+
2
|
| 21 |
+
],
|
| 22 |
+
"num_pooled_tokens": 1024,
|
| 23 |
+
"pad_input": true,
|
| 24 |
+
"patch_size": 14,
|
| 25 |
+
"processor_class": "KimiVLProcessor"
|
| 26 |
+
}
|
processor_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_map": {
|
| 3 |
+
"AutoProcessor": "processing_kimi_vl.KimiVLProcessor"
|
| 4 |
+
},
|
| 5 |
+
"processor_class": "KimiVLProcessor"
|
| 6 |
+
}
|
tokenizer_config.json
CHANGED
|
@@ -117,6 +117,7 @@
|
|
| 117 |
"<|media_pad|>"
|
| 118 |
],
|
| 119 |
"auto_map": {
|
|
|
|
| 120 |
"AutoTokenizer": [
|
| 121 |
"tokenization_moonshot.TikTokenTokenizer",
|
| 122 |
null
|
|
@@ -128,6 +129,7 @@
|
|
| 128 |
"extra_special_tokens": {},
|
| 129 |
"model_max_length": 1048576,
|
| 130 |
"pad_token": "[PAD]",
|
|
|
|
| 131 |
"tokenizer_class": "TikTokenTokenizer",
|
| 132 |
"unk_token": "[UNK]"
|
| 133 |
}
|
|
|
|
| 117 |
"<|media_pad|>"
|
| 118 |
],
|
| 119 |
"auto_map": {
|
| 120 |
+
"AutoProcessor": "processing_kimi_vl.KimiVLProcessor",
|
| 121 |
"AutoTokenizer": [
|
| 122 |
"tokenization_moonshot.TikTokenTokenizer",
|
| 123 |
null
|
|
|
|
| 129 |
"extra_special_tokens": {},
|
| 130 |
"model_max_length": 1048576,
|
| 131 |
"pad_token": "[PAD]",
|
| 132 |
+
"processor_class": "KimiVLProcessor",
|
| 133 |
"tokenizer_class": "TikTokenTokenizer",
|
| 134 |
"unk_token": "[UNK]"
|
| 135 |
}
|