Hugging Face's logo

LAXMAYDAY
/
FLUX.2-dev-int8-tensorwise

Text-to-Image
flux
comfyui
int8
quantized
Model card Files
xet
 Community
FLUX.2-dev-int8-tensorwise
File size: 2,643 Bytes
8f29560
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
{
  "artifact_target": "comfyui_diffusion_model",
  "config_path": null,
  "convrot": true,
  "convrot_groupsize": 256,
  "copied_tensor_count": 139,
  "cuda_max_memory_allocated_bytes": 2387263488,
  "cuda_max_memory_reserved_bytes": 4477419520,
  "dtype_counts": {
    "bfloat16": 139,
    "float32": 160,
    "int8": 160,
    "uint8": 160
  },
  "execution_device": "cuda:0",
  "missing_tensor_count": 0,
  "missing_tensors": [],
  "nonrotated_tensor_count": 0,
  "output_bytes": 33055815576,
  "output_checkpoint": "/workspace/models/export/flux2_int8_tensorwise.safetensors",
  "output_hash": "",
  "output_hash_state": "not_requested",
  "output_tensor_count": 619,
  "output_tensor_device": "cpu",
  "quant_metadata_tensor_count": 160,
  "quant_storage_dtype": "int8",
  "quantized_tensor_count": 160,
  "requested_device": "cuda:0",
  "rotated_tensor_count": 160,
  "scale_axis": "out_features",
  "scale_dtype": "fp32",
  "scale_granularity": "per_channel",
  "scale_tensor_count": 160,
  "schema_version": "int8_tensorwise_checkpoint_export_report.v1",
  "selected_source_files": {
    "flux2-dev.safetensors": 160
  },
  "source_checkpoint": "/workspace/models/hf-downloads/flux2/flux2-dev.safetensors",
  "source_file_count": 1,
  "source_format": "safetensors",
  "source_layout": "single_file",
  "source_tensor_count": 299,
  "status": "model_written",
  "target_dtype": "int8_tensorwise",
  "target_format": "safetensors",
  "tensor_index": {
    "format": {
      "bits": 8,
      "category": "integer_weight_activation",
      "name": "int8_tensorwise",
      "rounding": "nearest_even",
      "scale_axis": "out_features",
      "scale_granularity": "per_channel",
      "scale_method": "amax",
      "scale_required": true,
      "storage_dtype": "int8"
    },
    "schema_version": "quant_tensor_index.v1",
    "selection": {
      "algorithm": "int8_tensorwise",
      "algorithm_version": "0.1.0",
      "exclude": [],
      "include": [
        "double_blocks.*",
        "single_blocks.*"
      ],
      "keep_components": [
        "text_encoder",
        "vae"
      ],
      "kept_source_tensor_roles": [
        "bias"
      ],
      "quantized_module_count": 160,
      "quantized_tensor_count": 160,
      "source_tensor_roles": [
        "weight"
      ],
      "target_dtype": "int8_tensorwise"
    }
  },
  "written_files": [
    {
      "bytes": 33055815576,
      "hash": "",
      "hash_state": "not_requested",
      "kind": "int8_tensorwise_inference_checkpoint",
      "path": "/workspace/models/export/flux2_int8_tensorwise.safetensors",
      "state": "written",
      "tensor_count": 619
    }
  ]
}