Hugging Face's logo

labanochwo
/
unsloth-ocr-4bit

Image-to-Text
Transformers
Safetensors
English
qwen2_5_vl
text-generation-inference
unsloth
trl
4-bit precision
bitsandbytes
Model card Files
xet
Community
labanochwo commited on
Commit
dd24673
·
verified ·
1 Parent(s): 76e5d88

Upload model trained with Unsloth

Browse files

Upload model trained with Unsloth 2x faster

Files changed (5) hide show
  1. config.json +149 -0
  2. generation_config.json +14 -0
  3. model-00001-of-00002.safetensors +3 -0
  4. model-00002-of-00002.safetensors +3 -0
  5. model.safetensors.index.json +0 -0
config.json ADDED
@@ -0,0 +1,149 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2_5_VLForConditionalGeneration"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": 151643,
7
+ "eos_token_id": 151645,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 3584,
10
+ "image_token_id": 151655,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 18944,
13
+ "max_position_embeddings": 128000,
14
+ "max_window_layers": 28,
15
+ "model_type": "qwen2_5_vl",
16
+ "num_attention_heads": 28,
17
+ "num_hidden_layers": 28,
18
+ "num_key_value_heads": 4,
19
+ "pad_token_id": 151643,
20
+ "quantization_config": {
21
+ "bnb_4bit_compute_dtype": "float16",
22
+ "bnb_4bit_quant_type": "nf4",
23
+ "bnb_4bit_use_double_quant": true,
24
+ "llm_int8_enable_fp32_cpu_offload": false,
25
+ "llm_int8_has_fp16_weight": false,
26
+ "llm_int8_skip_modules": null,
27
+ "llm_int8_threshold": 6.0,
28
+ "load_in_4bit": true,
29
+ "load_in_8bit": false,
30
+ "quant_method": "bitsandbytes"
31
+ },
32
+ "rms_norm_eps": 1e-06,
33
+ "rope_scaling": {
34
+ "mrope_section": [
35
+ 16,
36
+ 24,
37
+ 24
38
+ ],
39
+ "rope_type": "default",
40
+ "type": "default"
41
+ },
42
+ "rope_theta": 1000000.0,
43
+ "sliding_window": 32768,
44
+ "text_config": {
45
+ "architectures": [
46
+ "Qwen2_5_VLForConditionalGeneration"
47
+ ],
48
+ "attention_dropout": 0.0,
49
+ "bos_token_id": 151643,
50
+ "eos_token_id": 151645,
51
+ "hidden_act": "silu",
52
+ "hidden_size": 3584,
53
+ "image_token_id": null,
54
+ "initializer_range": 0.02,
55
+ "intermediate_size": 18944,
56
+ "layer_types": [
57
+ "full_attention",
58
+ "full_attention",
59
+ "full_attention",
60
+ "full_attention",
61
+ "full_attention",
62
+ "full_attention",
63
+ "full_attention",
64
+ "full_attention",
65
+ "full_attention",
66
+ "full_attention",
67
+ "full_attention",
68
+ "full_attention",
69
+ "full_attention",
70
+ "full_attention",
71
+ "full_attention",
72
+ "full_attention",
73
+ "full_attention",
74
+ "full_attention",
75
+ "full_attention",
76
+ "full_attention",
77
+ "full_attention",
78
+ "full_attention",
79
+ "full_attention",
80
+ "full_attention",
81
+ "full_attention",
82
+ "full_attention",
83
+ "full_attention",
84
+ "full_attention"
85
+ ],
86
+ "max_position_embeddings": 128000,
87
+ "max_window_layers": 28,
88
+ "model_type": "qwen2_5_vl_text",
89
+ "num_attention_heads": 28,
90
+ "num_hidden_layers": 28,
91
+ "num_key_value_heads": 4,
92
+ "rms_norm_eps": 1e-06,
93
+ "rope_scaling": {
94
+ "mrope_section": [
95
+ 16,
96
+ 24,
97
+ 24
98
+ ],
99
+ "rope_type": "default",
100
+ "type": "default"
101
+ },
102
+ "rope_theta": 1000000.0,
103
+ "sliding_window": null,
104
+ "torch_dtype": "float16",
105
+ "use_cache": true,
106
+ "use_sliding_window": false,
107
+ "video_token_id": null,
108
+ "vision_end_token_id": 151653,
109
+ "vision_start_token_id": 151652,
110
+ "vision_token_id": 151654,
111
+ "vocab_size": 152064
112
+ },
113
+ "tie_word_embeddings": false,
114
+ "torch_dtype": "float16",
115
+ "transformers_version": "4.55.3",
116
+ "unsloth_version": "2025.8.9",
117
+ "use_cache": true,
118
+ "use_sliding_window": false,
119
+ "video_token_id": 151656,
120
+ "vision_config": {
121
+ "depth": 32,
122
+ "fullatt_block_indexes": [
123
+ 7,
124
+ 15,
125
+ 23,
126
+ 31
127
+ ],
128
+ "hidden_act": "silu",
129
+ "hidden_size": 1280,
130
+ "in_channels": 3,
131
+ "in_chans": 3,
132
+ "initializer_range": 0.02,
133
+ "intermediate_size": 3420,
134
+ "model_type": "qwen2_5_vl",
135
+ "num_heads": 16,
136
+ "out_hidden_size": 3584,
137
+ "patch_size": 14,
138
+ "spatial_merge_size": 2,
139
+ "spatial_patch_size": 14,
140
+ "temporal_patch_size": 2,
141
+ "tokens_per_second": 2,
142
+ "torch_dtype": "float16",
143
+ "window_size": 112
144
+ },
145
+ "vision_end_token_id": 151653,
146
+ "vision_start_token_id": 151652,
147
+ "vision_token_id": 151654,
148
+ "vocab_size": 152064
149
+ }
generation_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "attn_implementation": "flash_attention_2",
3
+ "bos_token_id": 151643,
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 151645,
7
+ 151643
8
+ ],
9
+ "max_length": 128000,
10
+ "pad_token_id": 151643,
11
+ "repetition_penalty": 1.05,
12
+ "temperature": 1e-06,
13
+ "transformers_version": "4.55.3"
14
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34c135bf428738a24ee82abaeea533e986db485395b961a87df46db860264464
3
+ size 4875747772
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4fafae458a69ec293eeff75bde071112e95cc53be5711bb3cfe8994a7fd43e1
3
+ size 1089994880
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff