---
base_model: Tongyi-MAI/Z-Image-Turbo
tags:
- orbitquant
- quantized
- diffusers
---

# Tongyi-MAI/Z-Image-Turbo OrbitQuant W2A3

This is an OrbitQuant artifact generated from the source model listed above.

## Quantization

- Method: `orbitquant`
- Bits: `W2A3`
- Runtime mode: `dequant_bf16`
- Activation kernel backend: `triton_cuda`
- Quantization device: `cuda`
- Weight quantization backend: `triton_cuda`
- Quantization staging: `component`
- Target policy: `z_image`
- Rotation: `rpbh`
- Rotation seed: `0`
- Block size: `paper`
- Block size policy: `largest_power_of_two_dividing_dim`
- Codebook: `lloyd_max`
- Codebook version: `1`
- Calibration data: none
- Text encoders and VAE: left in source precision by default

## Source

- Model: `Tongyi-MAI/Z-Image-Turbo`
- Revision: `f332072aa78be7aecdf3ee76d5c247082da564a6`
- Source license: `apache-2.0`
- OrbitQuant paper: https://arxiv.org/abs/2607.02461

## Limitations

The initial runtime may dequantize packed weights before BF16 matmul. Disk artifacts are compact; current CUDA/MPS kernels optimize selected activation codebook lookup/rescale stages and packed weight dequantization; full fused low-bit kernels are separate work.