Instructions to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5 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-V2.5:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5: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-V2.5:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5: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-V2.5:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5: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-V2.5:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5:Q4_K_M

Use Docker

docker model run hf.co/BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5:Q4_K_M

LM Studio
Jan
Ollama
How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5 with Ollama:
```
ollama run hf.co/BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5:Q4_K_M
```
Unsloth Desktop

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5 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-V2.5: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-V2.5:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5 with Docker Model Runner:
```
docker model run hf.co/BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5:Q4_K_M
```

Lemonade

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3-4b-Z-Image-Engineer-V2.5-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5 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-V2.5: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-V2.5:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5 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-V2.5: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-V2.5: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"

how to use this?

by manzomium - opened Jan 24

Discussion

manzomium

Jan 24

Pardon my ignorance, but I'm not sure how to use it.

I'm running ComfyUI on runpod. i uploaded the Q6 file in the tex-encoders folder, but COmy don't see it. Where do I need to save it?
I've also installed the custom node; Where do I need to connect it?

thanks a lot! I'd really like to use this as it sounds very promising

BennyDaBall

Owner Jan 25

•

edited Jan 25

Below is the default workflow I included in this repo:

You'll need City96 GGUF nodes or equivalent GGUF CLIP loader to use this as an enhanced clip model (which I highly recommend you try!). You should just be able to drop it into your text-encoders folder following installing a GGUF clip node. To use the LLM prompt enhancement feature you MUST run the model in LM Studio or another local host and configure the node appropriate for your setup (by default I have it setup for LM Studio implementation).

As a clip model just dropping it into you text encoders folder should do it once you have the City96 nodes installed (they come up under the bootleg category by default in custom node selector, or just double click and search GGUF CLIP and choose the appropriate loader).

On the VAE, any Z-image turbo compatible VAE (ae.safetensors, UltraFlux VAE, Z-Image_clear_vae, etc) will work.

scruffynerf

Jan 28

Benny, can you add the non-GGUF version to the repo? (Yes, I can manually build a safetensor version from the gguf, but it would be nice to get whatever originals in the highest quality (bf16 ?) you have, rather than a reconstruction.

BennyDaBall

Owner Jan 29

•

edited Jan 29

Yeah for sure! To be honest - some of it is a mess, I was manually transferring files between my mac and PC with external media and some of the base files may or may not be intact for the older versions as a result. So, it may take me a minute but I'll try to update when I get a chance!

BennyDaBall

Owner Jan 29

https://huggingface.co/BennyDaBall/Qwen3-4b-Z-Image-Engineer-V2.5/tree/main/safetensors_format

BennyDaBall changed discussion status to closed Jan 29

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