Skip to main content
Lightricks

Lightricks

Software companywww.lightricks.com

Search complete. 16 mentions across 9 episodes found for "Lightricks".

Sep 8, 2026

Alex VillabonHOST
9:14
What makes this acquisition interesting is the fact that if you look at what's happening, China's training models, the West's training models, like everyone, right? Like there's, you know, Black Forest Labs in Germany.
Alex VillabonHOST
9:26
There's Lightrix.
Alex VillabonHOST
9:30
Why Lightrix? They're somewhere here in Europe as well.
Alex VillabonHOST
9:32
I can't remember.
Alex VillabonHOST
9:35
Yeah,
Ofir BibiGUEST
1:11
I come from a research background and been at LTX for 11 years.
Ofir BibiGUEST
1:18
Well, it used to be Lightrix, now LTX for over 11 years.
Ofir BibiGUEST
1:24
And now kind of focusing on everything applications and partnerships and everything that we do around bringing LTX to the community and to our partners.
Doug HoganHOST
1:38
Yeah, yeah.
Addy GhaniHOST
50:50
Yeah.
Addy GhaniHOST
50:51
And then shout out to Black Forest Labs, as well as the LTX folks, Lightrix.
Addy GhaniHOST
50:57
This is obviously Stability AI.
Addy GhaniHOST
50:59
Yeah, there's a few smaller model companies left in the world.

Unknown podcast

Scoring the Cut: Building a Soundtrack Under Clips That Already Talk

Aug 23 · 1 Mention

speaker_0HOST
19:37
Alibaba shipped WAN 3.0 in August on top of the earlier 2.7 updates and is testing a Happy Horse 1.1 iteration alongside it.
speaker_0HOST
19:46
Lighttricks and partner platforms brought LTX 2.5 online in fast and pro tiers and Sand AI surfaced Magi 2 in preview, also open weights.
speaker_0HOST
19:56
If you have been meaning to bench an open weights model against your paid default, this is an unusually good month to do it.
speaker_0HOST
20:02
On the standings, and treat these strictly as this month's snapshot because they reshuffle every time we speak.

Unknown podcast

The Neural Daily 2026-08-20: GPU Banks and Rogue Agents

Aug 20 · 1 Mention

Roxanne StoneHOST
12:05
heavy.I suppose I'll have to upgrade Wyatt's desktop if I want to use it for our home security project.
Leonard CooperHOST
12:11
And let's not forget Lightricks LTX 2.5. They're using an asymmetric dual-stream diffusion transformer.
Leonard CooperHOST
12:21
They actually kept the audio and video streams separate, but coupled them with cross-attention.
Vector CaneHOST
12:27
It's a fidelity play.
speaker_0HOST
1:34
right? Yes, exactly.
speaker_0HOST
1:35
It's called LTX 2.5, and it was developed by LTX and Lightrix.
speaker_0HOST
1:40
We are going to explore how this specific technology is dragging AI video out of that novelty phase like we were just talking about and turning it into critical enterprise-grade infrastructure.
speaker_0HOST
1:51
Yeah,
Alex VolkovHOST
125:46
We talked about, uh, Cedens two point five finally releasing in the US, uh, Minimax H3, Flux three, right? Also video model, also open sourcing very s- hopefully very soon.
Alex VolkovHOST
125:57
And now we also have LTX from, from Lightricks.
Alex VolkovHOST
126:01
I think they spun out like a full...
Alex VolkovHOST
126:03
Correct me if I'm wrong, but I think they spun out like a company for LTX specifically.
Peter GostevGUEST
129:03
But yeah, it's, it's definitely, yeah, super, super cool to see.
Alex VolkovHOST
129:07
And fully open, which we absolutely love.
Alex VolkovHOST
129:09
So shout out to the Lightricks folks, LTX is their own company, research and models fully in the open.
Alex VolkovHOST
129:15
We really appreciate it.
HackerNoon AINARRATOR
0:00
This audio is presented by Hacker Noon, where anyone can learn anything about any technology.
HackerNoon AINARRATOR
0:05
LTX 2.5: A Complete Guide to Lightricks Audio-Video AI Model by AI Models forty-four.
HackerNoon AINARRATOR
0:13
Overview.
HackerNoon AINARRATOR
0:14
Is a diff-based audio-video foundation model built by Lightricks that generates synchronized video and audio in a single unified pass.
HackerNoon AINARRATOR
0:22
The model uses an asymmetric dual-stream transformer architecture with a 22B parameter distilled transformer for video generation and separate audio encoding, coupled through bi-directional audio-video cross-attention layers with modality-aware classifier-free guidance.
HackerNoon AINARRATOR
0:37
It accepts text prompts, images for image to video, or existing video clips as inputs and outputs MP4 files containing both video and synchronized audio.
HackerNoon AINARRATOR
5:20
Emerging feature maturity in spatial and audio quality.
HackerNoon AINARRATOR
5:24
While significantly improved from LTX 2, the distilled model retains much more rather than all of the full model's visual quality, indicating perceptible degradation.
Alon YarGUEST
13:18
Uh, and that was, like, maybe the most surprising moment, for me at least.
Alon YarGUEST
13:23
Uh, um, working at Lightricks for, like, eight years, coming from creative background and stuff, so robotics, it's, it's not something that I- I'm familiar with.
Alon YarGUEST
13:32
And suddenly seeing how they just use the same model that we just saw, like, on the VFX studio, that they use it for cinema, and they use the same foundational layer, the same infrastructure, but fine-tune it for their needs.
Alon YarGUEST
13:44
That was, like, pretty, uh, insane.

We value your privacy

We use cookies to understand how you use our platform and to improve your experience. Click “Accept All” to consent, or “Decline non-essential” to opt out of non-essential cookies. Read our Privacy Policy.