Content before more comment is regarded as page excerpt.

Until recently, the best way to generate images of a consistent character was from a trained lora. You would need to create a dataset of images and then train a FLUX lora on them.

In addition to making our FLUX.1 Kontext [dev] implementation open-source, we wanted to provide more guidance on how we chose to optimize it without compromising on quality.

Long time no see! We've been long overdue for an update.

And this one is really impressive. WE MADE AN ANIME OPENING!!! Er, I mean, Niji can now generate videos!!!!!!!!

While I'm generally psyched about each and every feature we make, I can say for sure that video-fying your pictures is the best damn thing we've shipped in a while. I really hope you enjoy it!

Late last week two great blog posts were released with seemingly opposite titles. "Don't Build Multi-Agents" by the Cognition team, and "How we built our multi-agent research system" by the Anthropic team.

Imagine a young girl named Emma who is fascinated by birds. Every weekend, she visits a nearby park to watch birds with her grandfather. Over time, Emma learns to recognize different bird species by their color, size, shape, and even their chirps. One afternoon, while flipping through a book, she effortlessly points to a picture and says, "Look, Grandpa! It's a robin!" She doesn't measure wingspans or analyze feather types; her brain instantly connects the image to her experiences and memories of robins at the park.

The conference, taking place in one of Europe's iconic art capitals, will feature a curated gallery that showcases how AI helps bring creative visions to life.

A Poetic Counterpart in Creation

When Paul Mouginot first brought AI into his art practice, he knew he'd tapped an exciting new tool for creative expression. But only through its continued use did he come to appreciate AI's versatility as an introspective entity — what he calls "a poetic counterpart in the act of creation."

Today, we are excited to release FLUX.1 Kontext, a suite of generative flow matching models that allows you to generate and edit images. Unlike existing text-to-image models, the FLUX.1 Kontext family performs in-context image generation, allowing you to prompt with both text and images, and seamlessly extract and modify visual concepts to produce new, coherent renderings.

Illustrious users often ask: “How can I get better results without writing long prompts?” Today, we’re excited to answer that with Text‑Enhancer, a new system that dramatically enriches user prompts for our image generation platform.

Text‑Enhancer actually comprises two intelligent components working together: