The world model the world
builds on just got better

LTX-2.5 is here. Smarter. Faster. Sharper. Still open. Still yours. Still LTX.

LTX-2.3 Examples Video

We build open world models that give you full control, from production-grade video to systems that understand and operate in the physical world.

//

LTX2.5 Model

New in LTX-2.5

New features, plus meaningful upgrades to what already worked.

01

Diffusion Fidelity Rendering

Render video a new way, building every scene from adaptive high-fidelity keyframes for pixel quality that holds up on a cinema screen.

ImageImage
ImageImage
Giant fluffy white yak with huge curved horns in a mountain valley with people and smaller yaks nearby.
02

Cleaner Motion

Generate smoother, more natural motion with fewer artifacts.

ImageImage
LTX-2.3 Rebuilt Latent Space
03

Native Multishot

Create connected shots that hold character, environment, lighting, and voice.

ImageImage
LTX-2.3 Better Prompt Adherence
04

Precise Control

Follow complex prompts, generate the right duration, and edit footage precisely.

ImageImage
LTX-2.3 Better Image To Video
05

Native 4K HDR

Generate native 4K HDR footage built for professional RAW and color pipelines.

ImageImage
LTX-2.3 Better Image To Video

06

More Efficient Generation

Get near-full-model quality with fewer retries and less compute.

ImageImage
LTX-2.3 Better Image To Video
//

Case Studies

One foundation.
Built to be put to work.

Studios, real-time platforms, and robotics teams all run production work on the same open model, extended for their own pipeline.

Physical AI

The open foundation robots train and run on

A pretrained foundation for deep adaptation across robotics, simulation, and embodied systems.

Training robots means teaching them how the physical world actually behaves, not just what it looks like. LTX-2.5 is the open model that gets closest to that for us, and being able to run and fine-tune it on our own hardware is what makes it usable for real robotics work. We have not found another open model that does this for us the way this one does.

real-time

The model fast enough to run live

Tune it to your own infrastructure to build avatars and products that respond in real time.

The future of media is real-time, and it only works if the underlying model is fast, controllable, and yours to run. Partnering with LTX means we can tune the model to our own infrastructure and bring live, responsive avatars to users worldwide.

creative production

From idea to pipeline in an afternoon

Native workflows, multishot continuity, and generative editing, all built for finishing-grade output.

ComfyUI and LTX share a simple belief, that open models build better tools, faster. With LTX-2.5 running natively in ComfyUI, a developer can go from an idea to a working world-model pipeline in an afternoon, and an enterprise can take that same pipeline straight into production on its own hardware. Open is what lets the community move fast and lets businesses build on what it proves.

creative production

A model you can actually control

Personalize it into new styles, characters, and capabilities, without losing control of the model itself.

We love building on top of LTX because it's one of the most capable models out there, and there's no other model we can personalize like this today. We already have over 80 fine-tunes of LTX on our platform, and that number grows every day. It's not a black box, it's something we can control, like a tool.

creative production

From Generation to Final Cut

Fine-tune it into your own pipeline to move from first cut to finishing-grade output.

Features like the ACES workflow are what excite us about LTX and its models. The LTX team is building genuine production workflows that integrate naturally into the hybrid environments we're creating at Asteria. LTX listens to filmmakers and delivers tools that work in real production. We can't wait to start using LTX 2.5.

//

Access

Yours to run.
Your way.

Run it on your own hardware. Fine-tune it on your own data. Keep everything you make.

Runs where you do

Any GPU. On-prem. Even a Mac. No render farm, no gatekeeper.

A fraction of the cost

Frontier-class output without the frontier bill.

Open weights, private fine-tuning

Your work stays yours. It is never rented back to you.

Free to build on

Free under $10M ARR. Ship first, talk later.

Experience It. Integrate It. Own It.

Try the model, build it into your product, or run the full stack yourself.