Yuvrajxms09
/

quant-FlashTalk

Model card Files Files and versions

quant-FlashTalk / fp8

37.8 GB

1 contributor

History: 3 commits

Yuvrajxms09's picture

Upload FP8 quantized FlashTalk DiT

fdcb56c verified 13 days ago

flashtalk_fp8_torchao.pt
Detected Pickle imports (14)
- "torch.storage.UntypedStorage",
- "torch._utils._rebuild_tensor_v3",
- "torchao.quantization.quantize_.common.kernel_preference.KernelPreference",
- "collections.OrderedDict",
- "torch.device",
- "torch._tensor._rebuild_from_type_v2",
- "torch.float8_e4m3fn",
- "torch.FloatStorage",
- "torch.serialization._get_layout",
- "torch.bfloat16",
- "torch.BFloat16Storage",
- "torch._utils._rebuild_wrapper_subclass",
- "torch._utils._rebuild_tensor_v2",
- "torchao.quantization.Float8Tensor"
How to fix it?
18.9 GB
xet

Upload FP8 quantized FlashTalk DiT 13 days ago
flashtalk_fp8_torchao.safetensors

18.9 GB
xet

Upload FP8 quantized FlashTalk DiT 13 days ago
flashtalk_fp8_torchao_meta.json

76 Bytes

Upload FP8 meta 13 days ago