codemichaeld/hy3d_2_turbo_fp8

FP8 Model Conversion

• —Source: https://huggingface.co/Comfy-Org/hunyuan3D_2.0_repackaged
• —Original File(s): hunyuan3d-dit-v2-mv-turbo_fp16.safetensors
• —Original Format: safetensors
• —FP8 Format: E4M3FN
• —FP8 File: hunyuan3d-dit-v2-mv-turbo_fp16-fp8-e4m3fn.safetensors

Usage

from safetensors.torch import load_file
import torch

# Load FP8 model
fp8_state = load_file("hunyuan3d-dit-v2-mv-turbo_fp16-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: 1649
• —Converted to FP8: 1649
• —Skipped (non-float): 0