134 lines
4.9 KiB
Markdown
134 lines
4.9 KiB
Markdown
|
---
|
||
|
|
license: apache-2.0
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
base_model:
|
||
|
|
- Tongyi-MAI/Z-Image-Turbo
|
||
|
|
library_name: gguf
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
tags:
|
||
|
|
- text-generation
|
||
|
|
- prompt-engineering
|
||
|
|
- image-generation
|
||
|
|
- z-image
|
||
|
|
- z-image-turbo
|
||
|
|
- qwen3
|
||
|
|
- gguf
|
||
|
|
- text-encoder
|
||
|
|
- comfyui
|
||
|
|
- lm-studio
|
||
|
|
- conversational
|
||
|
|
---
|
||
|
|
|
||
|
|
# Z-Image-Engineer V6 GGUF
|
||
|
|
|
||
|
|
**Follow me on X [@BennyDaBall_OG](https://x.com/BennyDaBall_OG) !**
|
||
|
|
|
||
|
|
GGUF quantized release for [Z-Image-Engineer V6](https://huggingface.co/BennyDaBall/Z-Image-Engineer-V6).
|
||
|
|
|
||
|
|
The main repo contains the merged HF safetensors. This repo contains the quant ladder for the [ComfyUI-Z-Engineer](https://github.com/BennyDaBall930/ComfyUI-Z-Engineer) node, LM Studio, ComfyUI `CLIPLoaderGGUF`, llama.cpp-style loaders, and local prompt-enhancement workflows.
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
## What is this?
|
||
|
|
|
||
|
|
**Z-Image-Engineer V6** is a SMART DoRA fine-tuned 4B Qwen text encoder from `Tongyi-MAI/Z-Image-Turbo`.
|
||
|
|
|
||
|
|
Use these GGUF files when you want:
|
||
|
|
|
||
|
|
- ComfyUI Z-Image text-encoder replacement and in-ComfyUI prompt enhancement through [ComfyUI-Z-Engineer](https://github.com/BennyDaBall930/ComfyUI-Z-Engineer) (no external server needed)
|
||
|
|
- LM Studio prompt enhancement
|
||
|
|
- ComfyUI text-encoder loading through plain `CLIPLoaderGGUF`
|
||
|
|
- smaller local files than the merged HF safetensors
|
||
|
|
- the same V6 prompt style and conditioning behavior in a quantized format
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
## Quantization Ladder
|
||
|
|
|
||
|
|
| Filename | Size | Target Use Case |
|
||
|
|
|---|---:|---|
|
||
|
|
| `Z-Image-Engineer-V6-F16.gguf` | 7.498 GiB | Full precision reference. |
|
||
|
|
| `Z-Image-Engineer-V6-Q8_0.gguf` | 3.986 GiB | Near-lossless; used for local A/B testing. |
|
||
|
|
| `Z-Image-Engineer-V6-Q6_K.gguf` | 3.079 GiB | High-fidelity balanced footprint. |
|
||
|
|
| `Z-Image-Engineer-V6-Q5_K_M.gguf` | 2.697 GiB | Daily-driver performance-to-size ratio. |
|
||
|
|
| `Z-Image-Engineer-V6-Q4_K_M.gguf` | 2.331 GiB | Reliable 4-bit standard. |
|
||
|
|
| `Z-Image-Engineer-V6-Q3_K_M.gguf` | 1.933 GiB | Lightweight option for tighter setups. |
|
||
|
|
| `Z-Image-Engineer-V6-MXFP4.gguf` | 2.101 GiB | Alternative compact quantization. |
|
||
|
|
|
||
|
|
Full recursive validation hashes are in `HASHES.sha256`.
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
## Quick Start
|
||
|
|
|
||
|
|
### LM Studio
|
||
|
|
|
||
|
|
Download a GGUF quant, load it, and prompt it directly:
|
||
|
|
|
||
|
|
```text
|
||
|
|
Enhance this image prompt for Z-Image Turbo: a unicorn
|
||
|
|
```
|
||
|
|
|
||
|
|
The comparison examples were generated from direct LM Studio user requests like this, with no separate system prompt. `V6_SYSTEM_PROMPT.md` is included only as an optional preset for people who want a stricter prompt-only chat setup.
|
||
|
|
|
||
|
|
### ComfyUI (recommended: ComfyUI-Z-Engineer)
|
||
|
|
|
||
|
|
1. Install the [ComfyUI-Z-Engineer](https://github.com/BennyDaBall930/ComfyUI-Z-Engineer) custom node (v2.0+).
|
||
|
|
2. Place a GGUF file into `ComfyUI/models/text_encoders/`.
|
||
|
|
3. Add **Z-Engineer CLIP Loader (GGUF)** and pick the quant - use the `clip` output where the stock Z-Image Qwen text encoder would normally go.
|
||
|
|
4. Optional: add **Z-Engineer Prompt Enhancer (Local)** with the same `clip` to rewrite seed prompts in-process, previewed on the node. No LM Studio or external server required.
|
||
|
|
|
||
|
|
A ready-made workflow ships with the node repo: `example_workflows/z_image_turbo_z_engineer.json`. With [ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF) installed the quant stays quantized in VRAM.
|
||
|
|
|
||
|
|
Alternative without the node: add a plain `CLIPLoaderGGUF` node (ComfyUI-GGUF), set model type to `lumina2`, and use it as the text encoder only.
|
||
|
|
|
||
|
|
Verified image settings:
|
||
|
|
|
||
|
|
```text
|
||
|
|
UNET: z_image_turbo_bf16.safetensors
|
||
|
|
VAE: ae.safetensors
|
||
|
|
Text Encoder: Z-Image-Engineer-V6-Q8_0.gguf
|
||
|
|
Resolution: 1024x1024
|
||
|
|
Steps: 8
|
||
|
|
CFG: 1.0
|
||
|
|
Sampler: res_multistep
|
||
|
|
Scheduler: simple
|
||
|
|
Shift: 3.0
|
||
|
|
```
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
## SMART DoRA
|
||
|
|
|
||
|
|
V6 was trained with BennyDaBall's SMART DoRA system:
|
||
|
|
|
||
|
|
- **DoRA** for direction/magnitude-separated adapter updates.
|
||
|
|
- **Entropic regularization** for less repetition and broader output variety.
|
||
|
|
- **Holographic regularization** for cleaner depth-wise feature structure.
|
||
|
|
- **Topological regularization** for more coherent latent trajectories.
|
||
|
|
- **Manifold regularization** for stable weight behavior during refinement.
|
||
|
|
|
||
|
|
The final V6 build used master-corpus SMART DoRA training, retention pressure, SceneClean SFT32 style restoration, AntiRepeat Binary24 refinement, and a 25% style-restoration / 75% anti-repeat DoRA blend.
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
## Related Repos
|
||
|
|
|
||
|
|
- Main merged HF release: [BennyDaBall/Z-Image-Engineer-V6](https://huggingface.co/BennyDaBall/Z-Image-Engineer-V6)
|
||
|
|
- ComfyUI custom node (GGUF + shard loaders, local prompt enhancer with preview): [ComfyUI-Z-Engineer](https://github.com/BennyDaBall930/ComfyUI-Z-Engineer)
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
## Acknowledgements
|
||
|
|
|
||
|
|
- **Tongyi-MAI** for the Z-Image Turbo ecosystem.
|
||
|
|
- **Qwen** for the adaptable text encoder backbone.
|
||
|
|
- The open-source maintainers behind **LM Studio**, **ComfyUI**, **llama.cpp**, **PEFT**, and **Transformers**.
|
||
|
|
|
||
|
|
**Built & trained locally with care by BennyDaBall.**
|
||
|
|
|
||
|
|
**Follow me on X [@BennyDaBall_OG](https://x.com/BennyDaBall_OG) !**
|