---
title: FLUX.2 Klein 4B CPU
emoji: 🎨
colorFrom: pink
colorTo: blue
sdk: gradio
sdk_version: 6.9.0
app_file: app.py
python_version: "3.11"
suggested_hardware: cpu-basic
startup_duration_timeout: 1h
preload_from_hub:
  - unsloth/FLUX.2-klein-4B-GGUF flux-2-klein-4b-Q4_K_M.gguf
  - Comfy-Org/vae-text-encorder-for-flux-klein-4b split_files/vae/flux2-vae.safetensors
short_description: Text-to-image and editing with FLUX.2 Klein on CPU
tags:
  - text-to-image
  - image-editing
  - flux
  - klein
  - cpu
  - gguf
license: apache-2.0
---

# FLUX.2 Klein 4B on Free CPU

Generate and edit images with FLUX.2 Klein 4B. Supports LoRA search from HuggingFace Hub.

Klein 4B is the smallest model that does both text-to-image and image editing in one model (Apache 2.0).

Upload a reference image + describe your edit, or just type a prompt to generate from scratch.

- Engine: stable-diffusion.cpp (GGUF Q4_K_M)
- Text encoder: Uncensored Qwen3-4B (Cordux)
- Steps: 4 (distilled) / Resolution: 512x512 default
- LoRA: Search and load any Klein 4B LoRA from HuggingFace
- Hardware: CPU Basic (2 vCPU, 16GB RAM)

## Credits

- [FLUX.2 Klein](https://bfl.ai/models/flux-2-klein) by Black Forest Labs
- [stable-diffusion.cpp](https://github.com/leejet/stable-diffusion.cpp) by leejet
- GGUF by [Unsloth](https://huggingface.co/unsloth)
- Uncensored encoder by [Cordux](https://huggingface.co/Cordux/flux2-klein-4B-uncensored-text-encoder)