Krea-2-Turbo is a text-to-image diffusion model hosted on Hugging Face. It generates images from text prompts and is provided under the krea-2-community-license.
The model is distributed in Safetensors format and is compatible with the Diffusers library. Users load it through a standard DiffusionPipeline after installing diffusers, transformers, and accelerate via pip. The recommended inference configuration uses bfloat16 precision on CUDA devices, with an option to switch to MPS for Apple hardware. Example code demonstrates loading the pipeline from the repository krea/Krea-2-Turbo and generating output for a detailed prompt describing a monochromatic blue halftone scene of a cat in an abandoned street.
It is made available by KREA. The repository includes a model card with instructions for local execution, notebooks, and inference providers. Deployment options encompass running the model directly in local applications or through compatible services on the Hugging Face platform. The English-language model card positions it within the class of text-to-image diffusion models.
Krea 2 Turbo is an Image generation project. It focuses on generating high-quality images from text prompts quickly using a specialized diffusion model. It is built as an open-source project for AI artists and developers. Krea 2 Turbo is open source under the Apache-2.0 license. It runs on the web, the command line, and API.
KREA builds and maintains Krea 2 Turbo, and it first shipped in 2026. The project is developed in the open on GitHub with 639 stars and 20 commits in the last 90 days. It operates in a well-populated space: PulseGate tracks 8 similar projects. Among its 3 catalogued features are text-to-image, diffusion pipeline, and high speed generation.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do