Krea
App : A product you use directly
Web studio bringing together the best image and video models, with real-time generation.
The best place to test multiple models without juggling multiple subscriptions.
Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.
What it does, concretely
Most image generation tools impose a repetitive cycle: write a prompt, wait, evaluate, repeat. Krea breaks this pattern by offering an immediate visual feedback loop, where the image builds in real time as you type or sketch.
In practice, Krea is a web-based studio that brings together over 64 image, video, and 3D models (Flux, Veo 3, Runway, Luma, Ideogram, as well as the in-house Krea 1 and Krea 2 models) into a single interface. Its real-time canvas renders results in under 50ms with each change. Generations are metered in compute units (credits) consumed per image, video, or upscale produced.
Krea is aimed at designers, architects, content creators, and creative teams who need to quickly test visual directions or maintain visual consistency across projects. The ability to train a custom model (LoRA) on a character, style, or product is especially useful for professionals working on recurring visual identities.
On the other hand, users who only need one specific model, or whose usage is very occasional, may find the credit system hard to predict: video generation, high-resolution upscaling, and LoRA training consume significantly more credits than simple image generation.
What it offers
- ·Real-time canvas: the image generates in under 50ms as you draw or type your prompt.
- ·Access to over 64 image, video, and 3D models (Flux, Veo 3, Runway, Luma, Ideogram, Krea 2) from a single interface, without managing multiple subscriptions.
- ·Image upscaling and enhancement up to 22,000 pixels resolution, with Topaz Photo AI and Topaz Gigapixel integration.
- ·Video generation and editing from text or images, with motion transfer between videos (paid).
- ·Custom LoRA model training on a character, style, or product to ensure visual consistency across generations (paid, full access from Pro plan).
- ·3D environments generated from an image or text via the Stage feature.
- ·Node Editor to chain multiple AI operations into a visual production pipeline (paid, full access from Pro plan).
- ·Free plan with 100 compute units per day, no credit card required.
In practice
- Price
- Freemium
- Commercial use
- Commercial use: paid plan
- Available on
- web
- FR interface
- No
Security and data
- Rating
- Security: one caveat, see the card
- Your data
- Data with third parties
- Maintenance
- Active
Links and repository checked on 25 October 2026 · added on 1 November 2023