Another SOTA model is here. Z-Image is a powerful and highly efficient image generation model built by Z-Image Team, Tongyi MAI, Alibaba Group with 6 billion parameters. It currently comes in three versions, and one of the most impressive among them is Z Image Turbo. This variant is a distilled, faster version of the main model and still manages to match and in some cases even beat the top competitors while using only 8 NFEs (Number of Function Evaluations).
 |
| Z-Image-Turbo showcase (Ref-official page) |
Z-Image-Turbo focuses heavily on photorealistic quality, producing sharp, clean visuals along with strong overall aesthetics. It also includes a Prompt Enhancer that adds reasoning capabilities, helping the model understand deeper context instead of just following surface-level descriptions. More detailed insights can be found into their research paper.
 |
| Z-Image-Turbo showcase (Ref-official page) |
This allows it to generate richer, more accurate results. Another standout feature is its ability to handle bilingual text extremely well.
 |
| Z-Image-Turbo showcase (Ref-official page) |
Installation
1. Install ComfyUI if you are new user. Update ComfyUI from the Manager by selecting Update All if you already using.
2. Download Z Image turbo released by the community(choose any of them as per system resources):
(a) Z Image turbo BF16 (z_image_turbo_bf16.safetensors) optimized by ComfyUI.
(b) Z Image turbo FP8 (z-image-turbo-fp8-e4m3fn.safetensors Or z-image-turbo-fp8-e5m2.safetensors) optimized by T5B. If you want better images then use E4M3FN. If you want maximum speed then use E5M2.
(c) Z Image turbo GGUF by Jayn7 (Q2 for fast inference to Q8 for better quality). Lower VRAM and system RAM users need to use the GGUF variants by analyzing its model size.
Save it into ComfyUI/models/diffusion_models folder.
For GGUF models, make sure you have ComfyUI-GGUF custom node by City 96. If not yet done, just install from Manager by selecting Custom Nodes Manager option. Update it if already using this.
If you do not know what is FP8/BF16/GGUF model variants, just follow our quantization tutorial to get more in depth overview.
3. Download Vae (ae.safetensors) and save this into ComfyUI/models/vae folder. This is the same vae that we use in Flux1. If you already have then downloading again is not required.
4. Download text encoder (qwen_3_4b.safetensors). Save this into ComfyUI/models/text_encoders folder.
5. Restart and refresh ComfyUI to take effect.
Workflow
1. Download the workflows from our Hugging Face repository.
- Z_Image_Turbo_Workflow.png (Basic BF16/FP8 workflow)
- Z_Image_Turbo (GGUF).json (GGUF variant workflow)
2. Drag and drop into ComfyUI.
(a) Load Z Image turbo fp8 on load diffusion model node.
(b) Load text encoders, vae into their respective node.
(c) Add positive and negative prompts into prompt box.
(d) Set KSampler Settings-
CFG- 0 (for turbo mode), normal use -1.0
Steps-9
Resolution -1024 by 1024
Sampler- Euler
(e) Hit run to start the execution.
Image Generation Testing
Prompt used- This is a realistic, analog-style photo of Karina. It captures a scene where she is attending a secret Illuminati party. She is winking, raising one hand above her head making a V-sign, and holding a highball glass in the other hand. The party features dazzling lasers and lights, and an LED screen displaying the Illuminati symbol. Aliens, Elon Musk, Donald Trump, and famous celebrities are dancing. The photo looks raw and unedited, characterized by visible film grain and the intense lighting from a flash. Cool lighting. Analog style. A candid, honest photo.
Prompt used-A stylish young woman standing confidently on a rainy New York street, wearing a fitted white tank top and a short red skirt. Her pose is elegant and natural, with one leg slightly forward and a relaxed yet confident expression. Reflections of city lights shimmer on the wet pavement around her. Yellow taxis, neon signs, and blurred pedestrians in the background create an authentic urban atmosphere. Raindrops gently fall, and her hair appears slightly damp from the rain. The overall scene is cinematic and photorealistic, with soft lighting, shallow depth of field, and a moody, vibrant color tone.
Prompt used- very pretty caucasian girl at age 18(with subtle alternative-style makeup and short, curly brown hair with soft layers and see-through side bangs), her hair is styled as a wolf cut, she is sitting in the selfie, which was taken at night in paris, with an average looking tenement visible in the background, its in the rural part of the city with a park in the back. The angle is messy, with slight motion blur and overexposure. The overall vibe is that of a casually taken, mediocre or even failed selfie — as if snapped without much thought or effort
){kind=link}