Sign In

MOSS-Image SFT Mixed INT8/W4 Quants

6

Updated: Nov 16, 2026

base modelmossint4int8int

Recently loaded

Download

1 variant available

int8 SafeTensor

moss_image_sft_int8_mixed.safetensors

8-bit integer, smaller file • 3.72 GB

Verified:

Type
Checkpoint Trained
Stats

58

6

Reviews
Published

Nov 16, 2026

Base Model
Hash
AutoV264992B6F5F
img_00085_.png

Quantization of MOSS-Image SFT aiming to reduce storage and RAM/VRAM.

This version uses a mix of INT8 and 4-bit weights and takes up approximately 3.72 GiB. It was created by testing different precision settings against the official INT8 model.

  • 107 layers use INT4 — W4A4 ConvRot MSE.

  • 61 layers use INT8 — the original INT8 ConvRot weights.

Workflow in image uses Lodestone's 8 Step Lora so my GPU won't scream for mercy

Tested on:

  • NVIDIA GTX 1660 Super 6 GB

  • 32 GB System RAM

Test settings:

  • 8/12 steps

  • CFG 1

  • Euler / Simple

  • 696 × 1048 resolution

  • Approximately 5.7 s/it on a GTX 1660 Super