FE2E-INT8

FE2E INT8 (Pre-quantized for CPU)

Pre-quantized INT8 model for FE2E (CVPR 2026) monocular depth + surface normal estimation from a single image.

Demo Space: WeReCooking2/FE2E-CPU

Files

Both files are saved with torch.save(model) (full model, not state_dict). Load with torch.load(..., mmap=True) to avoid doubling memory.

How it was made

1. Loaded FP32 base model (step1x-edit-i1258.safetensors) on GPU
2. Cast to FP32 on CPU
3. Merged LDRN LoRA in full precision
4. Applied torch.quantization.quantize_dynamic (INT8 on all nn.Linear layers)
5. Saved full model with torch.save(model)

Usage

import torch

dit = torch.load("dit_int8_full.pt", map_location="cpu", weights_only=False, mmap=True)
vae = torch.load("vae_full.pt", map_location="cpu", weights_only=False, mmap=True)

Requires ~12 GB RAM with mmap loading.

Performance

Single denoise step, outputs both depth and surface normal maps simultaneously.

No ONNX: PyTorch dynamo exporter produces a broken graph (100% NaN output).

Credits

• FE2E (CVPR 2026)
• Step1X-Edit base model
• rkfg/Step1X-Edit-FP8 FP8 quantization