Instructions to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

Use Docker

docker model run hf.co/BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

LM Studio
Jan

vLLM

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

SGLang

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with Ollama:
```
ollama run hf.co/BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M
```
Unsloth Desktop

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with Docker Model Runner:
```
docker model run hf.co/BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M
```

Lemonade

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3-4b-Z-Image-Engineer-V4-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "BennyDaBall/Qwen3-4b-Z-Image-Engineer-V4:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Could you please upload some images with prompts?

by junguang - opened Feb 9

Discussion

junguang

Feb 9

•

edited Feb 9

Hello, I'm a beginner. Could you please upload some images of ZITs created using this LLM model? I'm completely unfamiliar with photography terminology and would appreciate learning from the prompts on your images.

junguang changed discussion title from HOW USE "system prompt word" to Could you please upload some images with prompts? Feb 9

BennyDaBall

Owner Feb 9

•

edited Feb 9

The whole point is that it rewrites your prompts so you don't HAVE to know how to prompt well, or know any special terms, technical specs, or lighting terms. If you use a very simple input prompt it will give you a robust, properly formatted, enhanced single-paragraph prompt - experiment, have fun with it!

Here is an example - `Input prompt: "a portrait"` - Output from Z-Engineer: "A close-up shot of a person's face captured in soft focus against a deep black background, emphasizing intimate stillness. The subject is positioned slightly off-center, her gaze directed upward and to the right, creating an engaging connection with the viewer. Her skin exhibits fine organic texture—visible pores, faint lines around her eyes, and delicate creases at the corners of her mouth—all rendered in warm, natural tones that suggest a quiet life. A single strand of hair falls from the left side of the frame into soft focus, adding movement to the composition. The lighting is directional and gentle, originating from behind and slightly above, casting a luminous rim light along her forehead and cheekbones while leaving deep shadows in the hollows beneath her eyes and under her jawline. This creates strong contrast that enhances form without harshness. The background remains completely void of detail, allowing the subject’s features to dominate with quiet intensity. A subtle atmospheric haze softens the edges of distant elements beyond frame, reinforcing depth. The lens appears to be a normal focal length, providing natural perspective and minimal distortion while maintaining sharp focus on the face, capturing every nuance of expression in tactile realism. The overall mood is contemplative and serene, evoking a sense of quiet introspection through balanced lighting, rich texture, and layered spatial harmony."
- Image:

That said the real power comes from being able to set the Z-Engineer node to randomize seed (it will give you a new enhanced prompt every generation!) and set your ksampler seed to a seed you like (personal preference/experimentation required). That way you have infinite variation of your input prompt with a known good look - example:

stablecollector

Feb 23

example:

Hi I need to recreate my comfyui folder in stability matrix.
I got to your crux, but still to this day I don't understand and don't remember how I made the crux work last time.
I see from your image that you are using the model (I downloaded V4) with the z-Engineer node

I tried to install but it doesn't work, it always tells me the missing node in Comfyui.
Sorry for the inconvenience but could you give me a definitive solution to get your V4 model working as you are using it?

Thank you for your time.

BennyDaBall

Owner Feb 23

Hi I need to recreate my comfyui folder in stability matrix.
I got to your crux, but still to this day I don't understand and don't remember how I made the crux work last time.
I see from your image that you are using the model (I downloaded V4) with the z-Engineer node

I tried to install but it doesn't work, it always tells me the missing node in Comfyui.
Sorry for the inconvenience but could you give me a definitive solution to get your V4 model working as you are using it?

Thank you for your time.

Hey! I'm not familiar with Stability Matrix's setup so I might not be much help there... Sorry 😥 The Z-engineer models in GGUF can be used as CLIP text encoders using the City96 GGUF custom nodes. Again, I don't know anything about Stability Matrix but if it supports GGUF you should be able to use it as a CLIP!

To use this model with the Z-engineer custom node as an automatic-prompt-enhancer you need to be hosting the GGUF model in LMStudio locally.

stablecollector

Feb 24

To use this model with the Z-engineer custom node as an automatic-prompt-enhancer you need to be hosting the GGUF model in LMStudio locally.

Thanks for the reply.
I wanted to know exactly about the LMStudio locally issue.
Luckily I was able to find the custom node that also works in ComfyUI.

If anyone needs it, it's the "flybirdxx/ComfyUI-Prompt_Helper" repository. Read the GGUF model right away, without doing anything else.
With the 5070 TI it takes me 30 seconds with the v4 Q4.

Thanks again so much for your work.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment