Downloads · 30 days
26
20% of all-time downloads
lite-infer/z-image-turbo-nunchaku-lite-int4-r32
z-image-turbo-nunchaku-lite-int4-r32 is a text-to-image model from lite-infer. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as apache-2.0.
Downloads · 30 days
26
20% of all-time downloads
All-time downloads
133
Public
Parameters
3.3B
6.6 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors6.6 GB · 100%
How the weights are stored.
I83B · 91%
From the Hugging Face model README
Diffusers-loadable conversion of:
Tongyi-MAI/Z-Image-Turbonunchaku-ai/nunchaku-z-image-turbosvdq-int4_r32-z-image-turbo.safetensorsThe transformer uses quant_method: nunchaku_lite, INT4 SVDQ, group size 64, rank 32, and 238 quantized targets. Packed fused QKV and SwiGLU projections are split in logical layout into the stock Z-Image graph. The Qwen3 text encoder is packaged with BitsAndBytes 4-bit NF4.
| Checkpoint | Latency | Max VRAM |
|---|---|---|
| Converted Diffusers Nunchaku Lite INT4 r32 + BNB4 Qwen3 | 6.03 s (stdev 0.00 s) | 12.01 GiB |
RTX 5090, 1024×1024, 9 scheduler steps (8 DiT forwards), guidance scale 0, seed 42, one warmup and three measured runs. VRAM is peak total device usage sampled through nvidia-smi. The native row uses the base model's BF16 Qwen3 encoder.

Both images use the same prompt, seed, scheduler, resolution, and step count. Native Nunchaku 1.x selects NVFP4 kernels on Blackwell, so the comparison uses native NVFP4 as a visual reference only (cross-precision MAE 23.37, RMSE 30.81).
Requires the Hugging Face kernels package and an NVIDIA Turing, Ampere, Ada, or Blackwell (not Hopper) GPU.
import torch
from diffusers import ZImagePipeline
pipe = ZImagePipeline.from_pretrained(
"lite-infer/z-image-turbo-nunchaku-lite-int4-r32",
torch_dtype=torch.bfloat16,
).to("cuda")
image = pipe(
prompt='A cinematic portrait of a red fox in a misty forest at sunrise, detailed fur, volumetric light',
height=1024,
width=1024,
num_inference_steps=9,
guidance_scale=0.0,
generator=torch.Generator("cuda").manual_seed(42),
).images[0]
image.save("output.png")