gen‑ai.news
← Back
Image

Krea launched Krea 3 for fast and precise image editing

Krea launched Krea 3 for fast and precise image editing

Krea has launched Krea 3, its latest image editing model, promising 2K-resolution outputs in under 8 seconds. The company is positioning the model around both speed and adaptability, with the ability to be trained on a specific product, person, or visual style - making it a practical option for workflows that require consistent, repeatable results across many images.

The per-image pricing of $0.03 keeps the barrier to entry relatively low, particularly for commercial users who need to process large volumes of assets. That kind of usage-based pricing is increasingly common in the generative image space, where teams often prefer paying per output over flat subscription tiers, especially when workloads are variable.

The emphasis on under-8-second turnaround at 2K resolution is notable because that combination - quality and latency together - has historically been a trade-off in diffusion-based editing pipelines. Faster models have often meant lower fidelity, while high-resolution outputs have typically required longer processing times. If Krea 3 holds up at that speed in production conditions, it could fit comfortably into near-real-time creative workflows.

Krea has been building toward tighter iteration cycles for creative and commercial users, and Krea 3 fits that trajectory. The training capability for custom subjects or styles adds a personalization layer that goes beyond generic prompt-based editing, letting teams lock in a visual identity and apply it at scale. Details on the underlying architecture have not been widely published, but the product focus is clearly on practical throughput rather than purely on output quality benchmarks.

Enjoy this story? Get the next one in your inbox.

Twice a week: the most important stories in generative image and video AI, distilled into a 2-minute read.

Free. Unsubscribe any time. No spam, ever.

Your next read

FLUX 3 Image Klein: open weights, 5.6B parameters
Image

FLUX 3 Image Klein: open weights, 5.6B parameters

Black Forest Labs has released FLUX 3 Image Klein, a 5.6-billion-parameter open-weights model that brings frontier-level image quality alongside text-to-image generation, multi-reference generation, and editing capabilities. The release marks a notable step in making high-performing image models more accessible for customization and local deployment. Klein positions itself as a capable option for developers and researchers who want both output quality and the flexibility that open weights afford

Nobody is safe from celebrity AI deepfakes
Image

Nobody is safe from celebrity AI deepfakes

AI-generated images mimicking celebrities like Mia Goth and Taylor Swift are spreading across Instagram, sometimes through verified accounts with millions of followers. These deepfake posts exploit real-life events to appear credible, blurring the line between authentic celebrity content and fabricated material. The trend raises serious questions about platform accountability and the practical limits of content moderation at scale.