Hugging Face's logo

pytorch
/
Qwen3-8B-INT4

Text Generation
Transformers
PyTorch
multilingual
qwen3
torchao
code
math
chat
conversational
text-generation-inference
Model card Files
xet
Community
jerryzh168 commited on
Commit
a25e9e6
·
verified ·
1 Parent(s): fa3b0b5

Upload Qwen3ForCausalLM

Browse files
Files changed (5) hide show
  1. config.json +44 -4
  2. generation_config.json +1 -1
  3. pytorch_model-00001-of-00002.bin +2 -2
  4. pytorch_model-00002-of-00002.bin +2 -2
  5. pytorch_model.bin.index.json +1 -0
config.json CHANGED
@@ -11,6 +11,44 @@
11
  "hidden_size": 4096,
12
  "initializer_range": 0.02,
13
  "intermediate_size": 12288,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
  "max_position_embeddings": 40960,
15
  "max_window_layers": 36,
16
  "model_type": "qwen3",
@@ -18,13 +56,14 @@
18
  "num_hidden_layers": 36,
19
  "num_key_value_heads": 8,
20
  "quantization_config": {
21
- "include_embedding": false,
22
  "modules_to_not_convert": null,
23
  "quant_method": "torchao",
24
  "quant_type": {
25
  "default": {
26
  "_data": {
27
  "group_size": 128,
 
28
  "layout": {
29
  "_data": {
30
  "inner_k_tiles": 8
@@ -32,16 +71,17 @@
32
  "_type": "TensorCoreTiledLayout",
33
  "_version": 1
34
  },
 
35
  "preserve_zero": null,
36
  "set_inductor_config": true,
37
- "use_hqq": true,
38
  "zero_point_domain": {
39
  "_data": "NONE",
40
  "_type": "ZeroPointDomain"
41
  }
42
  },
43
  "_type": "Int4WeightOnlyConfig",
44
- "_version": 1
45
  }
46
  },
47
  "quant_type_kwargs": {},
@@ -53,7 +93,7 @@
53
  "sliding_window": null,
54
  "tie_word_embeddings": false,
55
  "torch_dtype": "bfloat16",
56
- "transformers_version": "4.52.0.dev0",
57
  "use_cache": true,
58
  "use_sliding_window": false,
59
  "vocab_size": 151936
 
11
  "hidden_size": 4096,
12
  "initializer_range": 0.02,
13
  "intermediate_size": 12288,
14
+ "layer_types": [
15
+ "full_attention",
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention"
51
+ ],
52
  "max_position_embeddings": 40960,
53
  "max_window_layers": 36,
54
  "model_type": "qwen3",
 
56
  "num_hidden_layers": 36,
57
  "num_key_value_heads": 8,
58
  "quantization_config": {
59
+ "include_input_output_embeddings": false,
60
  "modules_to_not_convert": null,
61
  "quant_method": "torchao",
62
  "quant_type": {
63
  "default": {
64
  "_data": {
65
  "group_size": 128,
66
+ "int4_choose_qparams_algorithm": "hqq",
67
  "layout": {
68
  "_data": {
69
  "inner_k_tiles": 8
 
71
  "_type": "TensorCoreTiledLayout",
72
  "_version": 1
73
  },
74
+ "packing_format": "tile_packed_to_4d",
75
  "preserve_zero": null,
76
  "set_inductor_config": true,
77
+ "use_hqq": false,
78
  "zero_point_domain": {
79
  "_data": "NONE",
80
  "_type": "ZeroPointDomain"
81
  }
82
  },
83
  "_type": "Int4WeightOnlyConfig",
84
+ "_version": 2
85
  }
86
  },
87
  "quant_type_kwargs": {},
 
93
  "sliding_window": null,
94
  "tie_word_embeddings": false,
95
  "torch_dtype": "bfloat16",
96
+ "transformers_version": "4.55.4",
97
  "use_cache": true,
98
  "use_sliding_window": false,
99
  "vocab_size": 151936
generation_config.json CHANGED
@@ -9,5 +9,5 @@
9
  "temperature": 0.6,
10
  "top_k": 20,
11
  "top_p": 0.95,
12
- "transformers_version": "4.52.0.dev0"
13
  }
 
9
  "temperature": 0.6,
10
  "top_k": 20,
11
  "top_p": 0.95,
12
+ "transformers_version": "4.55.4"
13
  }
pytorch_model-00001-of-00002.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c1cd775b881e92b5210c6deb171a6768f277907079ef65b0e2f94c6e72df326a
3
- size 4935497756
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e9cb30fa015b2d0075bc753865281b93b861b404b044bd852b3611ce0a4a020
3
+ size 4935466523
pytorch_model-00002-of-00002.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4fb405e09d822b45d676275cf6d4a22df256e0e21ac9be3d3e199b24610ab53e
3
- size 1244661125
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f41a2ea7d01f35ed3c0ecd8cab39da1cd7bc2ad48be7c46dc8fab80b5628bfa9
3
+ size 1244661564
pytorch_model.bin.index.json CHANGED
@@ -1,5 +1,6 @@
1
  {
2
  "metadata": {
 
3
  "total_size": 6179874816
4
  },
5
  "weight_map": {
 
1
  {
2
  "metadata": {
3
+ "total_parameters": 8190735360,
4
  "total_size": 6179874816
5
  },
6
  "weight_map": {