Instructions to use WeReCooking/flux2-klein-4B-uncensored-text-encoder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use WeReCooking/flux2-klein-4B-uncensored-text-encoder with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="WeReCooking/flux2-klein-4B-uncensored-text-encoder",
	filename="qwen3-4b-abl-q4_0.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use WeReCooking/flux2-klein-4B-uncensored-text-encoder with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0
# Run inference directly in the terminal:
llama-cli -hf WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0
# Run inference directly in the terminal:
llama-cli -hf WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0

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 WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0
# Run inference directly in the terminal:
./llama-cli -hf WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0

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 WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0

Use Docker

docker model run hf.co/WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0

LM Studio
Jan
Ollama
How to use WeReCooking/flux2-klein-4B-uncensored-text-encoder with Ollama:
```
ollama run hf.co/WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0
```

Unsloth Studio new

How to use WeReCooking/flux2-klein-4B-uncensored-text-encoder with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for WeReCooking/flux2-klein-4B-uncensored-text-encoder to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for WeReCooking/flux2-klein-4B-uncensored-text-encoder to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for WeReCooking/flux2-klein-4B-uncensored-text-encoder to start chatting

Pi new

How to use WeReCooking/flux2-klein-4B-uncensored-text-encoder with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use WeReCooking/flux2-klein-4B-uncensored-text-encoder with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0

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 WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0

Run Hermes

hermes

Docker Model Runner
How to use WeReCooking/flux2-klein-4B-uncensored-text-encoder with Docker Model Runner:
```
docker model run hf.co/WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0
```

Lemonade

How to use WeReCooking/flux2-klein-4B-uncensored-text-encoder with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull WeReCooking/flux2-klein-4B-uncensored-text-encoder:Q4_0

Run and chat with the model

lemonade run user.flux2-klein-4B-uncensored-text-encoder-Q4_0

List all available models

lemonade list

This is a duplicate of Cordux/flux2-klein-4B-uncensored-text-encoder for Flux-Klein-4B-CPU Space

Qwen3-4B Ablated (Uncensored) Text Encoder - GGUF Q4_0

Uncensored/ablated version of Qwen3-4B text encoder in GGUF Q4_0 format for Flux2 Klein 4B models.

Compatible Models

Flux2 Klein 4B (Distilled & Base)

What This Does

This is an ablated (safety-filtering removed) text encoder that allows Flux2 Klein models to generate NSFW content without prompt censorship.
The base Qwen3-4B text encoder that ships with Flux2 Klein has safety filtering that prevents certain prompts from being processed properly.

Installation

Download qwen3-4b-abl-q4_0.gguf
Place in ComfyUI/models/text_encoders/ or ComfyUI/models/unet/ (for GGUF loaders)
In your workflow, use a GGUF-compatible text encoder loader node
Point it to this file instead of the default Qwen3-4B

Prompting Tips

Use "wearing nothing" instead of "naked/nude" for best nude results
The model looks for clothing descriptors - even "nothing" counts as one
Clinical terms like "vagina" don't work better than colloquial terms
For explicit content beyond nudity, you'll need an NSFW LoRA

Language-Style Mapping Research

I discovered Flux.2 Klein associates languages with specific styles:
Japanese→anime portraits, German→illustrated art, etc.
Full study here

Limitations

This removes prompt filtering but doesn't add visual knowledge. The base Flux2 Klein models have limited training on explicit content, so:

✅ Nudity works well
✅ Suggestive poses work
❌ Explicit anatomy requires a LoRA
❌ Sexual acts require a LoRA

Credits

Based on huihui-ai/Qwen3-4B-abliterated
Converted with llama.cpp

Downloads last month: 2,005

GGUF

Model size

4B params

Architecture

qwen3

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for WeReCooking/flux2-klein-4B-uncensored-text-encoder

Base model

Qwen/Qwen3-4B-Base

Finetuned

Qwen/Qwen3-4B

Finetuned

huihui-ai/Qwen3-4B-abliterated

Quantized

(8)

this model

WeReCooking
/

flux2-klein-4B-uncensored-text-encoder