realrebelai/KREA-2_GGUFs overview
<p align="center" <a href="https://github.com/RealRebelAI" <img src="https://img.shields.io/badge/GitHub RealRebelAI E10600?style=for the badge&logo=github&log…
Runs locally from ~2.78 GB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| BASE/Krea-2-Base-Q3_K_M.gguf | GGUF | Q3_K_M | 5.17 GB | Download |
| BASE/Krea-2-Base-Q3_K_S.gguf | GGUF | Q3_K_S | 2.78 GB | Download |
| BASE/Krea-2-Base-Q4_K_M.gguf | GGUF | Q4_K_M | 6.76 GB | Download |
| BASE/Krea-2-Base-Q4_K_S.gguf | GGUF | Q4_K_S | 6.76 GB | Download |
| BASE/Krea-2-Base-Q5_K_M.gguf | GGUF | Q5_K_M | 8.26 GB | Download |
| BASE/Krea-2-Base-Q5_K_S.gguf | GGUF | Q5_K_S | 8.26 GB | Download |
| BASE/Krea-2-Base-Q6_K.gguf | GGUF | Q6_K | 9.86 GB | Download |
| BASE/Krea-2-Base-Q8_0.gguf | GGUF | Q8_0 | 12.76 GB | Download |
| TURBO/Krea-2-Turbo-Q3_K_M.gguf | GGUF | Q3_K_M | 5.14 GB | Download |
| TURBO/Krea-2-Turbo-Q3_K_S.gguf | GGUF | Q3_K_S | 5.14 GB | Download |
| TURBO/Krea-2-Turbo-Q4_K_M.gguf | GGUF | Q4_K_M | 6.72 GB | Download |
| TURBO/Krea-2-Turbo-Q5_K_S.gguf | GGUF | Q5_K_S | 8.21 GB | Download |
| TURBO/Krea-2-Turbo-Q6_K.gguf | GGUF | Q6_K | 9.80 GB | Download |
| TURBO/Krea-2-Turbo-Q8_0.gguf | GGUF | Q8_0 | 12.69 GB | Download |
Model Details
| Model ID | realrebelai/KREA-2_GGUFs |
|---|---|
| Author | realrebelai |
| Pipeline | — |
| License | other |
| Base model | krea/Krea-2-Turbo |
| Last modified | 2026-08-05T03:33:52.000Z |
Model README
---
license: other
license_name: krea-2-community-license
license_link: LICENSE
library_name: comfyui
base_model:
- krea/Krea-2-Turbo
base_model_relation: quantized
tags:
- comfyui
- gguf
- krea-2
---
<p align="center"> <a href="https://github.com/RealRebelAI"> <img src="https://img.shields.io/badge/GitHub-RealRebelAI-E10600?style=for-the-badge&logo=github&logoColor=E10600&labelColor=0D0D0D" alt="GitHub"> </a> <a href="https://huggingface.co/realrebelai"> <img src="https://img.shields.io/badge/Hugging%20Face-realrebelai-E10600?style=for-the-badge&logo=huggingface&logoColor=E10600&labelColor=0D0D0D" alt="Hugging Face"> </a> <a href="https://www.youtube.com/@RealRebelAI"> <img src="https://img.shields.io/badge/YouTube-%40RealRebelAI-E10600?style=for-the-badge&logo=youtube&logoColor=E10600&labelColor=0D0D0D" alt="YouTube"> </a> <a href="https://civitai.com/user/RealRebelAI"> <img src="https://img.shields.io/badge/Civitai-RealRebelAI-E10600?style=for-the-badge&logo=civitai&logoColor=E10600&labelColor=0D0D0D" alt="Civitai"> </a> <a href="https://x.com/realrebelai"> <img src="https://img.shields.io/badge/X-%40realrebelai-E10600?style=for-the-badge&logo=x&logoColor=E10600&labelColor=0D0D0D" alt="X"> </a> </p>
<h3 align="left">Krea 2 GGUF Sample Generations</h3>
<table align="center">
<tr>
<td><img src="https://cdn-uploads.huggingface.co/production/uploads/68761990332d15464ccc8dee/5L4UnkmiYc4fYOQsGc6CP.png" alt="Sample 1" width="400"/></td>
<td><img src="https://cdn-uploads.huggingface.co/production/uploads/68761990332d15464ccc8dee/Fk54nYMWLsi2tR7PHQJPN.png" alt="Sample 2" width="400"/></td>
</tr>
<tr>
<td align="center"><em></em></td>
<td align="center"><em></em></td>
</tr>
<tr>
<td><img src="https://cdn-uploads.huggingface.co/production/uploads/68761990332d15464ccc8dee/Wft1O2vOQhDvC7Jzfxj_l.png" width="400"/></td>
<td><img src="https://cdn-uploads.huggingface.co/production/uploads/68761990332d15464ccc8dee/eDyitcFvxGe1i1yX_cMPJ.png" width="400"/></td>
</tr>
<tr>
<td align="center"><em></em></td>
<td align="center"><em></em></td>
</tr>
</table>
KREA-2 GGUF Weights
Quantized GGUF weights for the Krea-2 Base model, optimized for low-VRAM and consumer hardware local workflows via ComfyUI.
🚀 Overview
Krea-2 is a Diffusion Transformer (DiT). Standard GGUF custom nodes will throw an Unexpected architecture type error natively. To run these weights, you must use our patched custom node fork.
📥 Downloading and Installation
1. Custom Node Requirements (Mandatory)
Standard loaders do not yet recognize krea2 architecture tags in GGUF metadata. Clone the following custom node into your ComfyUI/custom_nodes/ directory:
- Repository: RealRebelAI/ComfyUI-GGUF_KREA-2
2. Required Support Models
Place these base model files in your standard ComfyUI models/ directory:
- Text Encoder (CLIP): Qwen3-VL-4B-FP8-Scaled.safetensors (Load via
CLIPLoader, ensuring Type is set tokrea2) - VAE: qwen_image_vae.safetensors (Load via
VAELoader)
3. GGUF Weight Installation
- Download your preferred quantization file (e.g.,
Krea-2-Base-Q4_K_S.gguf) from the Files and versions tab of this repository. - Place the downloaded
.gguffile into yourComfyUI/models/unet/(or equivalent GGUF directory) folder. - Load the model in ComfyUI using the
UnetLoaderGGUFnode.
⚙️ Workflow Notes
Ensure your conditioning setup is mapped appropriately to handle the stacked hidden state requirements of multimodal DiTs to avoid latent sequence mismatch errors.
Run realrebelai/KREA-2_GGUFs with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models