doncarlos1/mio1

FP8 Model Conversion

• Source: https://huggingface.co/ponpoke/flux2-klein-9b-uncensored-text-encoder
• Original File(s): model.safetensors
• Original Format: safetensors
• FP8 Format: E4M3FN
• FP8 File: model-fp8-e4m3fn.safetensors

Usage

from safetensors.torch import load_file
import torch

# Load FP8 model
fp8_state = load_file("model-fp8-e4m3fn.safetensors")

# Convert tensors back to float32 for computation (auto-converted by PyTorch)
model.load_state_dict(fp8_state)

Note: FP8 tensors are automatically converted to float32 when loaded in PyTorch. Requires PyTorch ≥ 2.1 for FP8 support.

Statistics

• Total tensors: 399
• Converted to FP8: 399
• Skipped (non-float): 0