Z-Image-Turbo

Repository
Tongyi-MAI/Z-Image-Turbo
Engine
vLLM
Listed
31 Oct 2026

⚡️- Image An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer

Serves
Image generation
Parameters
4.0B
Weights
32.9 GB
License
Apache-2.0

Capabilities

Read from the model's capability tags. Requests are admitted per tag.

Image generation image_generation
  • POST /v1/images/generations
  • POST /v1/images/edits

Endpoint task mode image_generation

Inputs
Text

Launch settings

Every engine serving the model starts with these. An image or audio pipeline holds no token KV cache and does not decode token by token, so quantization, KV cache dtype and speculative decoding do not apply.

Set by the owner. auto: the serving node derives it. default: the platform's value, left in place.

Engine
vLLM
Scheduling policy --scheduling-policy
priority default
Interactive requests ahead of batch work; same-class traffic in arrival order.
Trust remote code --trust-remote-code
on default
Python shipped in the bundle may run at load (the default).

Engine tuning

1 of 24 set by the owner; the node derives the rest.

Set by the owner. auto: the serving node derives it.

Image generation

Profile
Curated profile
Limits read from the engine for this pipeline.
Images per request n
Set by the owner: 2
Input images image[]
0
Fixed sizes size
none (free sizes)
Smallest side size
256
Largest side size
2048
Side multiple size
16
Megapixel limit size
4
Default size size=auto
1024x1024
Inference steps num_inference_steps
9
Guidance scale guidance_scale
0
GPU memory floor GB per GPU
28.6623
Negative prompt negative_prompt
none
True CFG scale true_cfg_scale
none
Seed seed
none
Flow shift flow_shift
none
Second guidance scale guidance_scale_2
none
Edit strength strength
none

Memory and batching

GPU memory utilization --gpu-memory-utilization
auto
Max sequences --max-num-seqs
auto
Max batched tokens --max-num-batched-tokens
auto

Execution

Enforce eager --enforce-eager
auto
Async scheduling --no-async-scheduling
auto
Custom all-reduce --disable-custom-all-reduce
auto

Architecture

Read from the model's files at upload.

Architecture
ZImagePipeline
Parameters
4,022,468,096
Weights on disk
32.9 GB, safetensors

Call it

Create an endpoint for the model, then send it a request in the OpenAI shape.

Shell
# Create an endpoint for this model, then provision it. # --tier-id is the service tier to run on; pick one at /endpoints/new. erectl endpoints create \ --name "My endpoint" \ --slug my-endpoint \ --model-id 9ABED1C7-329C-41A0-95C6-D4B136455B72 \ --task-mode image_generation \ --tier-id <tier-id> # provision takes the endpoint UUID returned by create erectl endpoints provision <endpoint-id> # Call it. Requests are routed by /<project-id>/<endpoint-slug>. curl https://api.erebine.ai/proj_ABC123/my-endpoint/v1/images/generations \ -H "Content-Type: application/json" \ -H "Authorization: Bearer $EREBINE_API_KEY" \ -d '{ "model": "Z-Image-Turbo", "prompt": "A lighthouse at dusk, long exposure", "size": "1024x1024" }'

Model card

As published with Tongyi-MAI/Z-Image-Turbo. Images are left out; its HTML tables are set as tables.

&#160;
&#160;
&#160;
&#160;
&#160;
&#160;
&#160;
&#160;
&#160;

Welcome to the official repository for the Z-Image(造相)project!

✨ Z-Image

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.

📥 Model Zoo

ModelPre-TrainingSFTRLStepCFGTaskVisual QualityDiversityFine-TunabilityHugging FaceModelScope
Z-Image-Omni-Base✅❌❌50✅Gen. / EditingMediumHighEasyTo be releasedTo be released
Z-Image✅✅❌50✅Gen.HighMediumEasy
| | | Z-Image-Turbo | ✅ | ✅ | ✅ | 8 | ❌ | Gen. | Very High | Low | N/A | | | | Z-Image-Edit | ✅ | ✅ | ❌ | 50 | ✅ | Editing | High | Medium | Easy | To be released | To be released | | To be released |

🖼️ Showcase

📸 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.

🏗️ Model Architecture

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.

📈 Performance

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.

Click to view the full leaderboard

🚀 Quick Start

Install the latest version of diffusers, use the following command:
Click here for details for why you need to install diffusers from source

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 the latest official diffusers release.
Therefore, you need to install diffusers from source for the latest features and Z-Image support.

pip install git+https://github.com/huggingface/diffusers
import torch
from diffusers import ZImagePipeline

# 1. Load the pipeline
# Use bfloat16 for optimal performance on supported GPUs
pipe = ZImagePipeline.from_pretrained(
    "Tongyi-MAI/Z-Image-Turbo",
    torch_dtype=torch.bfloat16,
    low_cpu_mem_usage=False,
)
pipe.to("cuda")

# [Optional] Attention Backend
# Diffusers uses SDPA by default. Switch to Flash Attention for better efficiency if supported:
# pipe.transformer.set_attention_backend("flash")    # Enable Flash-Attention-2
# pipe.transformer.set_attention_backend("_flash_3") # Enable Flash-Attention-3

# [Optional] Model Compilation
# Compiling the DiT model accelerates inference, but the first run will take longer to compile.
# pipe.transformer.compile()

# [Optional] CPU Offloading
# Enable CPU offloading for memory-constrained devices.
# pipe.enable_model_cpu_offload()

prompt = "Young Chinese woman in red Hanfu, intricate embroidery. Impeccable makeup, red floral forehead pattern. Elaborate high bun, golden phoenix headdress, red flowers, beads. Holds round folding fan with lady, trees, bird. Neon lightning-bolt lamp (⚡️), bright yellow glow, above extended left palm. Soft-lit outdoor night background, silhouetted tiered pagoda (西安大雁塔), blurred colorful distant lights."

# 2. Generate Image
image = pipe(
    prompt=prompt,
    height=1024,
    width=1024,
    num_inference_steps=9,  # This actually results in 8 DiT forwards
    guidance_scale=0.0,     # Guidance should be 0 for the Turbo models
    generator=torch.Generator("cuda").manual_seed(42),
).images[0]

image.save("example.png")

🔬 Decoupled-DMD: The Acceleration Magic Behind Z-Image

Decoupled-DMD is the core few-step distillation algorithm that empowers the 8-step Z-Image model.

Our core insight in Decoupled-DMD is that the success of existing DMD (Distributaion Matching Distillation) methods is the result of two independent, collaborating mechanisms:

  • CFG Augmentation (CA): The primary engine 🚀 driving the distillation process, a factor largely overlooked in previous work.
  • Distribution Matching (DM): Acts more as a regularizer ⚖️, ensuring the stability and quality of the generated output.

By recognizing and decoupling these two mechanisms, we were able to study and optimize them in isolation. This ultimately motivated us to develop an improved distillation process that significantly enhances the performance of few-step generation.

🤖 DMDR: Fusing DMD with Reinforcement Learning

Building upon the strong foundation of Decoupled-DMD, our 8-step Z-Image model has already demonstrated exceptional capabilities. To achieve further improvements in terms of semantic alignment, aesthetic quality, and structural coherence—while producing images with richer high-frequency details—we present DMDR.

Our core insight behind DMDR is that Reinforcement Learning (RL) and Distribution Matching Distillation (DMD) can be synergistically integrated during the post-training of few-step models. We demonstrate that:

  • RL Unlocks the Performance of DMD 🚀
  • DMD Effectively Regularizes RL ⚖️

⏬ Download

pip install -U huggingface_hub
HF_XET_HIGH_PERFORMANCE=1 hf download Tongyi-MAI/Z-Image-Turbo

📜 Citation

If you find our work useful in your research, please consider citing:

@article{team2025zimage,
  title={Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer},
  author={Z-Image Team},
  journal={arXiv preprint arXiv:2511.22699},
  year={2025}
}

@article{liu2025decoupled,
  title={Decoupled DMD: CFG Augmentation as the Spear, Distribution Matching as the Shield},
  author={Dongyang Liu and Peng Gao and David Liu and Ruoyi Du and Zhen Li and Qilong Wu and Xin Jin and Sihan Cao and Shifeng Zhang and Hongsheng Li and Steven Hoi},
  journal={arXiv preprint arXiv:2511.22677},
  year={2025}
}

@article{jiang2025distribution,
  title={Distribution Matching Distillation Meets Reinforcement Learning},
  author={Jiang, Dengyang and Liu, Dongyang and Wang, Zanyi and Wu, Qilong and Jin, Xin and Liu, David and Li, Zhen and Wang, Mengmeng and Gao, Peng and Yang, Harry},
  journal={arXiv preprint arXiv:2511.13649},
  year={2025}
}