Omdano
/

INT8-H16P

Model card Files Files and versions

848 MB

Ctrl+K

Ctrl+K

1 contributor

History: 7 commits

Omdano's picture

Remove model.safetensors - use pytorch_model.bin with INT8 weights

29acac0 verified 7 months ago

.gitattributes

1.52 kB
initial commit 7 months ago
README.md

1.21 kB
Upload README.md with huggingface_hub 7 months ago
config.json

1.17 kB
Add back TorchAO INT8 quantization_config for proper loading 7 months ago
preprocessor_config.json

585 Bytes
Upload folder using huggingface_hub 7 months ago
pytorch_model.bin
Detected Pickle imports (15)
- "torch.device",
- "torchao.quantization.quant_primitives.ZeroPointDomain",
- "torch.BFloat16Storage",
- "torchao.dtypes.affine_quantized_tensor.AffineQuantizedTensor",
- "torch.CharStorage",
- "torch._utils._rebuild_tensor_v2",
- "torch._utils._rebuild_wrapper_subclass",
- "torch.int8",
- "torch.LongStorage",
- "torch._tensor._rebuild_from_type_v2",
- "torchao.dtypes.utils.PlainLayout",
- "torch.bfloat16",
- "collections.OrderedDict",
- "torch.serialization._get_layout",
- "torchao.dtypes.uintx.plain_layout.PlainAQTTensorImpl"
How to fix it?
848 MB
xet

Upload folder using huggingface_hub 7 months ago