Krea 2 Raw
Base 12B Krea 2 checkpoint for finetuning, post-training, and open-weight text-to-image workflows
Krea 2 Raw
Base 12B Krea 2 checkpoint for finetuning, post-training, and open-weight text-to-image workflows
Krea 2 Raw Overview
Krea 2 Raw is the base release checkpoint in the Krea 2 model family, published as an open-weight 12B text-to-image model before the additional post-training and distillation used for Krea 2 Turbo. Krea describes it as a better fit for finetuning, post-training, and custom domain work than for general inference, making it especially relevant for LoRA training and other open-model customization workflows.
More models from Krea
Krea 2 Turbo is the open-source Krea image endpoint exposed on Runware for fast, low-cost text-to-image generation. It focuses on prompt-driven generation rather than the richer Krea 2 reference workflow, making it a better fit for simple production pipelines, accelerated runtimes, and workflows that benefit from open-model compatibility such as LoRA-based customization.
Krea 2 Medium Turbo is the fast Krea 2 variant that keeps the richer Krea creative workflow around moodboards, creativity tuning, and reference-driven generation. It is designed for rapid ideation and iteration-heavy image work while still supporting image-to-image generation, style references, and the broader Krea 2 control surface that teams use for guided exploration.
Krea 2 Medium is the smaller, faster, and more cost-efficient variant in the Krea 2 family. It applies heavier post-training for especially stable and consistent outputs, making it a strong fit for illustration, anime, painting, graphic design, and other expressive visual workflows. It supports text-to-image and image-to-image generation, prompt interpretation strength through the creativity control, and up to 10 weighted reference images with both positive and negative guidance.
Krea 2 Large is the higher-capacity variant in the Krea 2 family, with lighter post-training and a more textured, flexible output character. It is the stronger overall model when a workflow benefits from higher ceiling, stronger photorealism, more raw visual character, and better handling of motion blur, film grain, low dynamic range, and other less polished looks. It supports text-to-image and image-to-image generation, prompt interpretation strength through the creativity control, and up to 10 weighted reference images with both positive and negative guidance.
![FLUX.1 Krea [dev]](/_next/image?url=https%3A%2F%2Fassets.runware.ai%2Fcovers%2Fbfl-flux-1-krea-dev.jpg&w=3840&q=75)
FLUX.1 Krea [dev] is an open-weight text-to-image model from Black Forest Labs and Krea AI. It targets opinionated aesthetics and realistic photography. Developers can drop it into FLUX.1 dev workflows to build custom generators that avoid the typical AI look.