Model reference · open weights
Z-Image is an open-weight image model from Alibaba. Z-Image-Turbo (FP32) weighs 16.4 GB; the smallest configuration that runs it is RTX 4090 24 GB.
Summary of the Tongyi-MAI/Z-Image-Turbo model card, 2026-10-01
What it is
| Released by | Alibaba |
|---|---|
| Released | 2025-11-25 |
| Parameters | 6.2B |
| VRAM | 16.4 GB for the weights |
What it runs on
| Card | One image | Memory |
|---|---|---|
| RTX 3060 12 GB … RTX 4060 Ti 16 GB | does not fit | |
| RTX 3090 24 GB | tight | 23.4 GB |
| RTX 4090 24 GB | tight | 23.4 GB |
| RTX 5090 32 GB | fits | 31.0 GB |
| L40S 48 GB | fits | 44.0 GB |
| A100 80 GB | fits | 78.2 GB |
| H100 80 GB | fits | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | fits | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | fits | 107 GB |
| H200 141 GB | fits | 138 GB |
| B200 180 GB | fits | 176 GB |
Builds
| Build | Params | Precision | Weights | Smallest setup |
|---|---|---|---|---|
| Z-Image-Turbo (above) ↗ | 6.2B | FP32 | 16.4 GB | RTX 4090 24 GB |
| Z-Image ↗ | 6.2B | BF16 | 20.5 GB | RTX 4090 24 GB |
| Z-Image-Turbo-GGUF ↗ packaged by unsloth 15 builds: Q2_K 3.6 GB … F16 12.3 GB the transformer alone, to use with the text encoders and VAE |
— | GGUF | 5.0 GB | RTX 3060 12 GB |
| Z-Image-Turbo ↗ packaged by unsloth |
6.2B | FP32 | 16.4 GB | RTX 4090 24 GB |
| Z-Image-Turbo-unsloth-bnb-4bit ↗ packaged by unsloth |
3.2B | INT4 | 6.8 GB | RTX 3060 12 GB |
From the model card
Welcome to the official repository for the Z-Image（造相）project!
Z-Image is a powerful and highly efficient image generation model family with 6B parameters. Currently there are four variants:
🚀 Z-Image-Turbo – A distilled version of Z-Image that matches or exceeds leading competitors with only 8 NFEs (Number of Function Evaluations). It offers ⚡️sub-second inference latency⚡️ on enterprise-grade H800 GPUs and fits comfortably within 16G VRAM consumer devices. It excels in photorealistic image generation, bilingual text rendering (English & Chinese), and robust instruction adherence.
🎨 Z-Image – The foundation model behind Z-Image-Turbo. Z-Image focuses on high-quality generation, rich aesthetics, strong diversity, and controllability, well-suited for creative generation, fine-tuning, and downstream development. It supports a wide range of artistic styles, effective negative prompting, and high diversity across identities, poses, compositions, and layouts.
🧱 Z-Image-Omni-Base – The versatile foundation model capable of both generation and editing tasks. By releasing this checkpoint, we aim to unlock the full potential for community-driven fine-tuning and custom development, providing the most "raw" and diverse starting point for the open-source community.
✍️ Z-Image-Edit – A variant fine-tuned on Z-Image specifically for image editing tasks. It supports creative image-to-image generation with impressive instruction-following capabilities, allowing for precise edits based on natural language prompts.
📸 Photorealistic Quality: Z-Image-Turbo delivers strong photorealistic image generation while maintaining excellent aesthetic quality.
📖 Accurate Bilingual Text Rendering: Z-Image-Turbo excels at accurately rendering complex Chinese and English text.
💡 Prompt Enhancing & Reasoning: Prompt Enhancer empowers the model with reasoning capabilities, enabling it to transcend surface-level descriptions and tap into underlying world knowledge.
🧠 Creative Image Editing: Z-Image-Edit shows a strong understanding of bilingual editing instructions, enabling imaginative and flexible image transformations.
We adopt a Scalable Single-Stream DiT (S3-DiT) architecture. In this setup, text, visual semantic tokens, and image VAE tokens are concatenated at the sequence level to serve as a unified input stream, maximizing parameter efficiency compared to dual-stream approaches.
According to the Elo-based Human Preference Evaluation (on Alibaba AI Arena), Z-Image-Turbo shows highly competitive performance against other leading models, while achieving state-of-the-art results among open-source models.
Install the latest version of diffusers, use the following command:
We have submitted two pull requests (#12703 and #12715) to the 🤗 diffusers repository to add support for Z-Image. Both PRs have been merged into th
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.
Running it yourself
Rent a machine by the hour. ComfyUI is installed on it. Open ComfyUI through the tunnel, then Templates → Z-Image-Turbo: Text to Image: it loads ComfyUI's own build of this model, the files above.
# on your rented machine: pip install diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained("Tongyi-MAI/Z-Image-Turbo", torch_dtype=torch.bfloat16).to("cuda")
image = pipe(prompt="a red bicycle on a cobbled street").images[0]
image.save("/workspace/out.png")# on your rented machine (the ssh line is on its page in the console)
# get REPO FILE FOLDER: one file into /workspace/models/FOLDER, where ComfyUI loads it from
get() { hf download "$1" "$2" --local-dir /workspace/hf-files && mkdir -p "/workspace/models/$3" && mv "/workspace/hf-files/$2" "/workspace/models/$3/$4"; }
# the files of the template “Z-Image-Turbo: Text to Image” (19.3 GB)
get Comfy-Org/z_image_turbo split_files/text_encoders/qwen_3_4b.safetensors text_encoders
get Comfy-Org/z_image_turbo split_files/vae/ae.safetensors vae
get Comfy-Org/z_image_turbo split_files/diffusion_models/z_image_turbo_bf16.safetensors diffusion_models
start-comfyui# on your computer, in a second terminal: ComfyUI in your browser at http://localhost:8188
# HOST and PORT are your machine's, from its page in the console
ssh -L 8188:localhost:8188 dev@HOST -p PORT