Z Image Turbo needs about 4.8 GB of memory at Q4_K (a 3.9 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 7 quants, from 2.6 GB (Q2_K) to 6.6 GB (Q8_0).
6B‑parameter image generation model with efficient single‑stream diffusion
Z Image Turbo is a 6.5 B parameter image generation model. It is a distilled version of Z‑Image, offering fast inference and strong photorealistic output. The model supports bilingual text rendering and instruction‑following in English and Chinese.
Good at
- Fast inference on consumer GPUs
- High‑quality photorealistic images
- Accurate bilingual text rendering
Limits
- No vision support
- Requires 8 k context for memory estimates
- License is apache‑2.0
Best for
- Creative image generation
- Fine‑tuning for artistic styles
- Image editing with natural language prompts
Who it’s for. Developers and artists who need a lightweight image generation model that runs on consumer‑grade GPUs. The model is suitable for projects requiring bilingual text rendering and instruction adherence.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 2.6 GB | 3.4 GB | Fits | Get file → | |
| Q3_K | 3.1 GB | 4.0 GB | Fits | Get file → | |
| Q4_0 | 3.7 GB | 4.6 GB | Fits | Get file → | |
| Q4_K | 3.9 GB | 4.8 GB | Fits | Get file → | |
| Q5_0 | 4.5 GB | 5.6 GB | Fits | Get file → | |
| Q6_K | 5.3 GB | 6.4 GB | Fits | Get file → | |
| Q8_0 | 6.6 GB | 7.9 GB | Tight | Get file → |
For 4 GB VRAM use Q2_K (2.59 GB). For 8 GB VRAM use Q4_K (3.86 GB). For 16 GB VRAM use Q6_K (5.26 GB). Memory estimates are approximate and based on 8 k context.
What GPU do you need for Z Image Turbo?
These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is Q3_K (3.1 GB file, about 4.0 GB in use).
- 6 GB VRAM: the best quant that fits is Q5_0 (4.5 GB file, about 5.6 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (6.6 GB file, about 7.9 GB in use).
Which Z Image Turbo quant should you choose?
Q4_K (3.9 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
If memory is tight, the smallest file is Q2_K at 2.6 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Z Image Turbo
How much VRAM does Z Image Turbo need?
About 4.8 GB at Q4_K with an 8K context window. That is the 3.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Z Image Turbo on an 8 GB GPU?
Yes, with Q8_0: the 6.6 GB file needs about 7.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Z Image Turbo GGUF?
It depends on the quant: Q2_K is 2.6 GB, Q3_K is 3.1 GB, Q4_0 is 3.7 GB, Q4_K is 3.9 GB, Q5_0 is 4.5 GB, Q6_K is 5.3 GB and 1 more.
Where can I download Z Image Turbo GGUF?
The files are on Hugging Face in the leejet/Z-Image-Turbo-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Z Image Turbo use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Z Image Turbo on 4 GB VRAM?
Yes, the Q2_K quantized file is 2.59 GB and fits within 4 GB VRAM with some overhead.
What context length does the model support?
The model is evaluated at an 8 k token context length for memory estimates.
Is the model licensed for commercial use?
The model is released under the apache‑2.0 license, which permits commercial use.