Introducing the Leonardo Agent. Start with one idea. Build it into a character, a real-world scene, band photography, album art and ads—while keeping the identity intact. Meet Inference and their EP, Mortal Koi. We created the full world one prompt at a time with the agent. Available at no extra cost, for a limited time. Generations use your existing tokens. Follow along.

Different ideas, different creations. 👀 See what the community is making with the Leonardo Agent. Which one is your favorite?

One idea. Every asset. Byte Tidy is a brand made with the Leonardo Agent. One brief became a full campaign: illustration, photography, localized ads and social cuts. What would you brief it on?

A walk through spaces the community built with the Leonardo Agent, one mood at a time. Which one would you step into?

Nano Banana 2.1 is here 🍌 Include all your ideas: who's in the scene, the framing, the texture of every surface. It better understands and follows even the most detailed prompts, so more of your vision comes through. Available now in Leonardo.

Ideogram 4.5 is live in Leonardo, and we're excited to be a launch partner 🎉 It's the most precise image editing model yet from @ideogram_ai, made for multi-step editing. Reposition products, swap palettes, translate labels, or reshoot at golden hour. What's working stays put, and the rest is yours to play with. Try it now on Leo.

Design at Scale is coming 📷 Our own Joel Perlgut is joining Beau Penton from Canva for the third episode of Canva for Creatives — a live, under-the-hood look at how AI-powered workflows are changing professional design. Joel will be talking through how @Affinity, Leonardo.Ai and @canva fit together in a real creative pipeline — from high-end design work to scaling it across a whole team. 📷 24 September 📷 Register: bit.ly/4AHWPCO Know a designer who needs to see this? Tag them or send it their way.

How much control do you actually have over AI video? More than you think. We broke down a few of the techniques behind Seedance Tribal, a short film made with Seedance 2.5 in Leonardo. From getting scale right, to knowing when to give the model room to surprise you.

Some stories are best told without a script. Made by our creative team as an experiment with Seedance 2.5 in Leonardo, Seedance Tribal is a short film with no dialogue or conventional plot - just costume, color and movement carrying the story forward. It explores what happens when you treat an AI video model not as a shortcut, but as a filmmaking tool. Watch it now and stay tuned as we share some of the techniques that brought it to life. #MadeWithLeonardo

The model you already love, one level up. ✨ Wan 3.0 has arrived, making video creation easier, sharper and more controllable. Available now on Leo.

Voiceover, without the studio booking. Seed Audio turns your script into narration that actually fits the moment, tone and pacing included. Good for tutorials, walkthroughs, and any story worth hearing out loud. Try it in Leonardo.

Thirty-second shots with native audio. Up to fifty references — your characters, your world, your sounds, your directions — all kept consistent across every frame. Seedance 2.5, now in Leonardo. Built for directing, not guessing.

FLUX 3 Video: Picture and sound, made in one pass. Black Forest Labs' first move into video has arrived. Generate video and audio in a single pass, up to 20 seconds per clip. Great for: - Achieving consistent characters and identity across chained multi-shot sequences. - Cinematic and documentary-style aesthetic. - Dialogue support across multiple languages. Built for short narrative and atmospheric video. Try it on Leo, now.

MiniMax H3 is on Leonardo. Most video models give you a clip. H3 gives you the clip and the soundtrack — with your character and voice locked in from refs. Expect: - Commercial-grade quality across ads, e-commerce, gaming & UI - Best value in its class - One lightweight model, fully multimodal Built for one-shot finished content: brand teasers, product videos, fashion films, talking characters