FP32 Trained QWEN VL Text Encoder & FP32 Trained VAE

This model has gradient estimation/smoothing using Montecarlo for the diffuison model

Model Overview

Model Name: Krea 2
Version: v1.0
Release Date: June 22, 2026
Model Type: Text-to-image diffusion model
Architecture: Diffusion Transformer with 12 billion parameters
License: Krea 2 Community License
Release Format: Open-weight release and Krea-hosted product integrations
Model Developer: Krea.ai, Inc.

Downloads last month: 36

Safetensors

Model size

13B params

Tensor type

F32

Model tree for Felldude/Krea-2-Raw

Base model

krea/Krea-2-Raw

Finetuned

(30)

this model