Krea

5 indexed models

5 matching models

Krea: K3 Edit

krea/k3-edit-20260914

Krea K3 Edit is an image-to-image editing model from Krea. It takes one to eight reference images plus a text prompt and returns a single edited image, following instructions that range from small local changes to complex multi-step edits while keeping the subjects and compositio

Context window
65,536
Input modalities
text, image
Output modalities
image
Capabilities
imageOutput, multimodal, visionInput

Krea: Krea 2 Large

krea/krea-2-large-20260720

Krea 2 Large is Krea's high-capability image generation model, more than twice the size of Krea 2 Medium. Its lighter post-training gives images a rawer, more textured, and flexible character, making it particularly suited for photorealism, expressive artistic styles, motion blur

Context window
65,536
Input modalities
text, image
Output modalities
image
Capabilities
imageOutput, multimodal, visionInput

Benchmarks

  • LMArena: 1,113.971

Krea: Krea 2 Medium

krea/krea-2-medium-20260720

Krea 2 Medium is Krea's balanced, cost-efficient image generation model and a practical starting point for a broad range of use cases. Its extensive post-training supports stable, consistent generations, with particular strengths in illustration, anime, painting, and other expres

Context window
65,536
Input modalities
text, image
Output modalities
image
Capabilities
imageOutput, multimodal, visionInput

Benchmarks

  • LMArena: 1,128.139

Krea: Krea 2 Medium Turbo

krea/krea-2-medium-turbo-20260720

Krea 2 Medium Turbo is a distilled, speed-focused variant of Krea 2 Medium from Krea. It is designed for rapid iteration and graphic design exploration where fast generation is the priority.

Context window
65,536
Input modalities
text, image
Output modalities
image
Capabilities
imageOutput, multimodal, visionInput

krea: Krea 3

cloudflareWorkersAI/krea/krea-3

Krea 3 is Krea's image generation and editing model, in beta. It edits one or more reference images from a text instruction, or generates an image from the prompt alone. References are passed as HTTP(S) URLs in image_urls (up to 8). Outputs go up to 2K, with optional thinking_mod

Input modalities
image, text
Output modalities
image
Capabilities
imageOutput, multimodal, visionInput

Models

Loading
EVIDENCE / COMPARISON

Benchmarks

Explore every benchmark

Loading

These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

Install TensorTrace