🖼️ POCKET-Image-Zimage — 4-bit (NF4) Z-Image for on-device

A 4-bit (NF4) quantized build of Z-Image (Apache-2.0), packaged by VIDRAFT for low-VRAM, on-device image generation — part of the POCKET line.

• 📦 ~6 GB on disk (transformer + text encoder in NF4, VAE in fp16)
• ⚡ Runs from ~8.6 GB VRAM (≈4.5 GB with CPU offload) — vs 23.3 GB for bf16
• 🎯 ~2.7–5× smaller footprint, quality on par with the bf16 base

Usage

import torch
from diffusers import ZImagePipeline   # or ZImageImg2ImgPipeline / ZImageInpaintPipeline

pipe = ZImagePipeline.from_pretrained(
    "FINAL-Bench/POCKET-Image-Zimage", torch_dtype=torch.bfloat16
).to("cuda")
img = pipe("a serene mountain lake at sunrise, photorealistic", num_inference_steps=20).images[0]
img.save("out.png")

Requires bitsandbytes (CUDA). Measured reload + generate peak: 10.9 GB VRAM. For Apple Silicon / CPU, an optimum-quanto int8 build (13.4 GB) is the portable option.

🎨 The full POCKET-Image system

This repo hosts the quantized base model only. The headline character-perfect Korean & multilingual text feature is delivered by the POCKET-Image pipeline, not by these weights alone. Try the full system here:

• 🎨 Studio (generate here): https://huggingface.co/spaces/FINAL-Bench/POCKET-Image-Studio

Base model: Tongyi-MAI/Z-Image (Apache-2.0) · Quantization: bitsandbytes NF4 · By VIDRAFT.

---

🧩 The POCKET Family — On-device AI by VIDRAFT

Big models, small hardware. No GPU, no cloud.

Models

• 📦 POCKET-35B-GGUF — flagship, PC / server, no GPU
• 📦 POCKET-26B-GGUF — compact 26B
• 🇰🇷 POCKET-KR-GGUF — Korean, Android
• 🍎 POCKET-KR-MLX — Korean, iPhone / Mac
• 🌍 POCKET-EN-GGUF — English, phone / PC
• 🖼️ POCKET-Image-Zimage — 4-bit Z-Image (this repo)

Demos & tools (Spaces)

• 🎨 POCKET-Image Studio — text-in-image, generate in-page
• 🖥️ POCKET-35B-CPU — 35B answering on a CPU
• 🖥️ POCKET-26B-CPU — 26B on a CPU

📚 Full POCKET collection