Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

LTX HDR beta is now live. Every AI video model before this one output 8-bit SDR only. Fine for social clips. The format falls apart the moment you try to grade. Highlights clip. Shadows crush. AI footage won't composite cleanly against higher-bit-depth CGI. Resolution was never the real issue.... show more

LTX

14,233 subscribers

25,137,221 views • 4 months ago •via X (Twitter)

Arts Education Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Kling AI has impressed me so much with the new Kling 2.0 that I've made a special compilation video to showcase my work so far. 2.0 has the best prompt adherence of any model I've experienced. This is the future! 1.6 was already good, but this is the model I've been waiting for. Everything below is a one-try output. Kling AI

Kling AI has impressed me so much with the new Kling 2.0 that I've made a special compilation video to showcase my work so far. 2.0 has the best prompt adherence of any model I've experienced. This is the future! 1.6 was already good, but this is the model I've been waiting for. Everything below is a one-try output. Kling AI

WuxIA Rocks

14,680 views • 1 year ago

higgsfield DROPPED this and it just made ai filmmaking easier than ever. they open-sourced their originals. every flagship video they produced in-house, the entire prompt is now public. the references, the generation settings, the model, the resolution, all of it sitting right there on the project page. for anyone making ai video this is the part that actually matters. the hardest thing was never the tool, it was knowing how to prompt it. what to write to get cinematic lighting, character consistency across shots, real physics, camera movement that doesn't look ai. that knowledge stayed locked inside studios and a few people who figured it out early. now it's just open. open one of their best clips, read the exact prompt that built it, copy it, download the video, rebuild the whole thing yourself.

higgsfield DROPPED this and it just made ai filmmaking easier than ever. they open-sourced their originals. every flagship video they produced in-house, the entire prompt is now public. the references, the generation settings, the model, the resolution, all of it sitting right there on the project page. for anyone making ai video this is the part that actually matters. the hardest thing was never the tool, it was knowing how to prompt it. what to write to get cinematic lighting, character consistency across shots, real physics, camera movement that doesn't look ai. that knowledge stayed locked inside studios and a few people who figured it out early. now it's just open. open one of their best clips, read the exact prompt that built it, copy it, download the video, rebuild the whole thing yourself.

Sulfur

31,111 views • 1 month ago

The winners won't be AI experts. They'll be domain experts with AI agents. Here's how it plays out: - Phase 1: Being digitally native was the advantage - Phase 2: Being AI native was the advantage - Phase 3: Using AI Agents is the advantage ← we are here - Phase 4: Building agents that critique your agents The move right now: - Make an agent that loops against your output. - Before anything publishes, it asks: "Is this a 9/10 on these metrics?" If not, it rewrites. Tests. Learns. AI skills get commoditized. AI systems compound. I'm rooting for you. Go build the system.

The winners won't be AI experts. They'll be domain experts with AI agents. Here's how it plays out: - Phase 1: Being digitally native was the advantage - Phase 2: Being AI native was the advantage - Phase 3: Using AI Agents is the advantage ← we are here - Phase 4: Building agents that critique your agents The move right now: - Make an agent that loops against your output. - Before anything publishes, it asks: "Is this a 9/10 on these metrics?" If not, it rewrites. Tests. Learns. AI skills get commoditized. AI systems compound. I'm rooting for you. Go build the system.

The Startup Ideas Podcast (SIP) 🧃

12,249 views • 7 months ago

$2,000 to $10,000 off the World Cup from football clips that were never filmed, generated from one sentence The World Cup is feeding 3 billion eyeballs into short-form video right now, and she's filling that demand with footage no camera ever shot > Claude writes the prompt, an AI playground generates a 10-second clip > it looks like real broadcast footage, no stadium, no crew > under 3 minutes per clip, posted as Shorts > rip real footage and it gets claimed, generate it and you keep every dollar 3 billion people are searching, the real highlights can't meet that she's not competing with the broadcasters, she's filling the gap they can't, with clips nobody can claim

$2,000 to $10,000 off the World Cup from football clips that were never filmed, generated from one sentence The World Cup is feeding 3 billion eyeballs into short-form video right now, and she's filling that demand with footage no camera ever shot > Claude writes the prompt, an AI playground generates a 10-second clip > it looks like real broadcast footage, no stadium, no crew > under 3 minutes per clip, posted as Shorts > rip real footage and it gets claimed, generate it and you keep every dollar 3 billion people are searching, the real highlights can't meet that she's not competing with the broadcasters, she's filling the gap they can't, with clips nobody can claim

Woody

13,366 views • 2 months ago

I’ve been using Topaz Labs for my video upscaling needs, but choosing the right model depends on what your footage needs. Starlight Precise 2.5 is the detail machine. It’s ideal for sharpening real or older footage, or refining AI video that already has strong structure but needs more clarity and fine detail. Astra 2 is the fixer. I use it for AI-generated footage that needs correction, especially clips with warped details, motion artifacts, flicker, blur or that slightly plastic AI look. Here’s a before-and-after comparison showing both models taking 720p footage to 4K. Two different tools, two different jobs. Be sure to watch in full screen to fully appreciate the difference. Try both at

I’ve been using Topaz Labs for my video upscaling needs, but choosing the right model depends on what your footage needs. Starlight Precise 2.5 is the detail machine. It’s ideal for sharpening real or older footage, or refining AI video that already has strong structure but needs more clarity and fine detail. Astra 2 is the fixer. I use it for AI-generated footage that needs correction, especially clips with warped details, motion artifacts, flicker, blur or that slightly plastic AI look. Here’s a before-and-after comparison showing both models taking 720p footage to 4K. Two different tools, two different jobs. Be sure to watch in full screen to fully appreciate the difference. Try both at

Captain HaHaa

14,983 views • 2 months ago

I wanted to see if AI could handle story, not just visuals. with Kling 3.0, I planned multiple shots, locked the character identity and let the engine handle motion and transitions. I still did a final manual pass, color grading, subtle SFX just to make it fully post ready. but honestly, the raw output was already clean. very minimal glitches, no weird deformations and the movement feels noticeably smoother than previous models. The only downside is generation time, it’s a bit slower but that’s understandable when everyone’s using a new release. Overall, this is the most solid AI video model I’ve tested so far. One more thing worth mentioning: Higgsfield AI 🧩 provides unlimited generations of Kling 3.0 with 70% OFF. If you’re building real sequences, not just clips, you should try it Kling AI #higgsfieldpartner

I wanted to see if AI could handle story, not just visuals. with Kling 3.0, I planned multiple shots, locked the character identity and let the engine handle motion and transitions. I still did a final manual pass, color grading, subtle SFX just to make it fully post ready. but honestly, the raw output was already clean. very minimal glitches, no weird deformations and the movement feels noticeably smoother than previous models. The only downside is generation time, it’s a bit slower but that’s understandable when everyone’s using a new release. Overall, this is the most solid AI video model I’ve tested so far. One more thing worth mentioning: Higgsfield AI 🧩 provides unlimited generations of Kling 3.0 with 70% OFF. If you’re building real sequences, not just clips, you should try it Kling AI #higgsfieldpartner

DStudioproject

80,205 views • 7 months ago

Want to create your own AI Scott Adams video? Today we’re releasing an early beta that lets anyone generate a ~15 second AI Scott clip on any topic. Copy/paste a headline. Watch AI Scott respond.Use it for your own curiosity or drop it into an X conversation and let AI Scott do the talking . Again this is sneak peak beta of something much bigger we are working on. Short clips only for now. And not the full AI model used in daily AICWSA yet. Yes rendering is slow. We’ll improve as capacity allows. If demand explodes, we'll throttle off before we cause a GPU brownout. Elon’s space data centers can’t get here fast enough. The goal of this AI project is to realize the wishes of Scott Adams that he discussed here: And here’s the video on how to access the beta now:

Want to create your own AI Scott Adams video? Today we’re releasing an early beta that lets anyone generate a ~15 second AI Scott clip on any topic. Copy/paste a headline. Watch AI Scott respond.Use it for your own curiosity or drop it into an X conversation and let AI Scott do the talking . Again this is sneak peak beta of something much bigger we are working on. Short clips only for now. And not the full AI model used in daily AICWSA yet. Yes rendering is slow. We’ll improve as capacity allows. If demand explodes, we'll throttle off before we cause a GPU brownout. Elon’s space data centers can’t get here fast enough. The goal of this AI project is to realize the wishes of Scott Adams that he discussed here: And here’s the video on how to access the beta now:

AI Coffee with Scott Adams

52,179 views • 6 months ago

Today, history has been made. Veo 3 image-to-video is now here – a landmark moment in the history of moving image art. Just like film was revolutionized by the introduction of sound, generative AI has made leaps with this groundbreaking model. Try it for yourself in FLORA.

Today, history has been made. Veo 3 image-to-video is now here – a landmark moment in the history of moving image art. Just like film was revolutionized by the introduction of sound, generative AI has made leaps with this groundbreaking model. Try it for yourself in FLORA.

FLORA ©

16,478 views • 1 year ago

What if AI could finally reveal your Social Capital and put it to work? Gigi v2 is live. We spent the last decades optimizing for Content. The AI era will be about Context. For the first time, AI can surface social data that was invisible. It can map the value of your network, measure your credibility, and read you reputation, which are the foundations of something we’ve never been able to quantify until now: Social Capital. The only currency that drives access, influence, and opportunities in life. Because in a world where AI does the work, the only edge left is who you are, and how people see you. New paradigm. New protocol. New status game. Let’s play. 👉 Join the beta (link in comments) 💬 Follow me and Comment to skip the line 🔄 Repost for super priority access

What if AI could finally reveal your Social Capital and put it to work? Gigi v2 is live. We spent the last decades optimizing for Content. The AI era will be about Context. For the first time, AI can surface social data that was invisible. It can map the value of your network, measure your credibility, and read you reputation, which are the foundations of something we’ve never been able to quantify until now: Social Capital. The only currency that drives access, influence, and opportunities in life. Because in a world where AI does the work, the only edge left is who you are, and how people see you. New paradigm. New protocol. New status game. Let’s play. 👉 Join the beta (link in comments) 💬 Follow me and Comment to skip the line 🔄 Repost for super priority access

Clara Gold

98,962 views • 1 year ago

This channel has 20 million views on one skeleton short No face, no voice, and at the short form rate that pays about $900 I scored 1,593 uploads from 29 real AI channels against their own medians A hit means beating your own channel by 10x The base rate is 6.91 percent - one upload in fourteen Short form hits 6.43 percent and pays cents per thousand views Long form hits 9.54 percent and pays dollars Same million views, two orders of magnitude apart, decided before you render a frame Animal drama is the most copied format in AI video - 459 attempts IT CONVERTS AT 0.60x THE BASE RATE History scene, 212 attempts, runs 2.12x $12 to generate one, one in fourteen lands Everyone publishes the winner, nobody publishes the other thirteen The pipeline was never the hard part The format you pick before the first render is Full breakdown below

This channel has 20 million views on one skeleton short No face, no voice, and at the short form rate that pays about $900 I scored 1,593 uploads from 29 real AI channels against their own medians A hit means beating your own channel by 10x The base rate is 6.91 percent - one upload in fourteen Short form hits 6.43 percent and pays cents per thousand views Long form hits 9.54 percent and pays dollars Same million views, two orders of magnitude apart, decided before you render a frame Animal drama is the most copied format in AI video - 459 attempts IT CONVERTS AT 0.60x THE BASE RATE History scene, 212 attempts, runs 2.12x $12 to generate one, one in fourteen lands Everyone publishes the winner, nobody publishes the other thirteen The pipeline was never the hard part The format you pick before the first render is Full breakdown below

Skaly_Bull

15,267 views • 29 days ago

Been using AI video tools for 2 years. My real complaint was never quality, it was the grind. Generate, hate it, generate again, sometimes a dozen times before something's usable. Seedance Enterprise 2.5 gave me a usable result on generation for this one. That's the difference that actually matters. Made this on BytePlus Lumina, with Seedream and Seed Audio feeding the references. Seedance 2.5 API is about to launch on BytePlus, and now you can try it directly on Lumina.

Been using AI video tools for 2 years. My real complaint was never quality, it was the grind. Generate, hate it, generate again, sometimes a dozen times before something's usable. Seedance Enterprise 2.5 gave me a usable result on generation for this one. That's the difference that actually matters. Made this on BytePlus Lumina, with Seedream and Seed Audio feeding the references. Seedance 2.5 API is about to launch on BytePlus, and now you can try it directly on Lumina.

Gina Acosta

57,106 views • 1 month ago

Say goodbye to traditional journaling. 👋 Session Buddy beta is live... and it's free for 2 months The first-ever AI trading note-taker that: → Records your thought process in real time → Tracks every decision you make → Logs all your data automatically Your journal can't do this. Try it now:

Say goodbye to traditional journaling. 👋 Session Buddy beta is live... and it's free for 2 months The first-ever AI trading note-taker that: → Records your thought process in real time → Tracks every decision you make → Logs all your data automatically Your journal can't do this. Try it now:

TradePath

67,625 views • 6 months ago

I'm a man of my word, since you guys really needed it spelled out this way: Is this a 3D model? called us out so now we have to add the following disclaimer to this post: This is not a 3D model or gameplay footage in any regard. It is merely a dumb AI generated video clip. This post was sponsored by SocialKettle. Would you play a game like this?

I'm a man of my word, since you guys really needed it spelled out this way: Is this a 3D model? called us out so now we have to add the following disclaimer to this post: This is not a 3D model or gameplay footage in any regard. It is merely a dumb AI generated video clip. This post was sponsored by SocialKettle. Would you play a game like this?

Morph

68,674 views • 8 months ago

We made a short film that asks one question: "what is reality?" For a year, every AI film we made answered it for us: you could still tell. Then BytePlus handed us Seedance 2.0 in native 4K, and we couldn't. Filmmakers have a word for footage clean enough to ship: final pixel. AI video has never been there, because resolution was always the ceiling. This is the first time the frame held everything we asked of it. It finally looks the way we imagined it. "Reality is our workshop. Imagination is our tool." Watch REALITY in 4K, then try it yourself: BytePlus Lumina

We made a short film that asks one question: "what is reality?" For a year, every AI film we made answered it for us: you could still tell. Then BytePlus handed us Seedance 2.0 in native 4K, and we couldn't. Filmmakers have a word for footage clean enough to ship: final pixel. AI video has never been there, because resolution was always the ceiling. This is the first time the frame held everything we asked of it. It finally looks the way we imagined it. "Reality is our workshop. Imagination is our tool." Watch REALITY in 4K, then try it yourself: BytePlus Lumina

Amir D

306,455 views • 2 months ago

Did you know that arcade cabinets with CRT monitors actually had three different resolutions? Most classic 2D arcade games used a 15kHz (320×240) resolution. This was the same resolution commonly used by CRT TVs and many 8-bit, 16-bit, and 32-bit home consoles. Then Sega introduced higher-resolution arcade games with Virtua Fighter and Virtua Racing, which used 24kHz (496×384). Sega Model 1, 2, and 3 arcade boards are famous for using this resolution. One of the reasons the Sega Saturn versions of Virtua Fighter 2 and Daytona USA looked so different from the arcade versions was because of this resolution gap. Later 3D arcade games moved to 31kHz (640×480) resolution. Sega NAOMI games are one of the best-known examples. That’s why you need to be careful when buying an arcade cabinet. Early cabinets only supported 15kHz, while the commonly seen Astro City supported both 15kHz and 24kHz. Later cabinets like Blast City, Net City, and New Net City were known as “3-mode” cabinets because they supported all three resolutions. The tricky part is that both the game board and the arcade cabinet itself can have resolution settings. If the settings are wrong, you can end up running a high-resolution game in low resolution. So if you’re buying an arcade cabinet, make sure you carefully check which resolutions it supports.

Did you know that arcade cabinets with CRT monitors actually had three different resolutions? Most classic 2D arcade games used a 15kHz (320×240) resolution. This was the same resolution commonly used by CRT TVs and many 8-bit, 16-bit, and 32-bit home consoles. Then Sega introduced higher-resolution arcade games with Virtua Fighter and Virtua Racing, which used 24kHz (496×384). Sega Model 1, 2, and 3 arcade boards are famous for using this resolution. One of the reasons the Sega Saturn versions of Virtua Fighter 2 and Daytona USA looked so different from the arcade versions was because of this resolution gap. Later 3D arcade games moved to 31kHz (640×480) resolution. Sega NAOMI games are one of the best-known examples. That’s why you need to be careful when buying an arcade cabinet. Early cabinets only supported 15kHz, while the commonly seen Astro City supported both 15kHz and 24kHz. Later cabinets like Blast City, Net City, and New Net City were known as “3-mode” cabinets because they supported all three resolutions. The tricky part is that both the game board and the arcade cabinet itself can have resolution settings. If the settings are wrong, you can end up running a high-resolution game in low resolution. So if you’re buying an arcade cabinet, make sure you carefully check which resolutions it supports.

Gamer_lafan

73,847 views • 3 months ago

World Cup 2026 + Claude + AI clips = $100k a month on autopilot FIFA made YouTube a preferred platform for the tournament, so the search is pouring in, and the real highlights can't keep up > real match footage gets claimed by Content ID and demonetized > AI-generated clips, a packed stadium, flags, a slow-motion goal, belong to no one and keep every dollar > sports CPMs are up 20 to 40% during the tournament > Claude writes the clip, a generator builds it in 3 minutes, no real frame in any of it She's 22 and has never watched a match the edge isn't knowing soccer, it's being on the right side of Content ID while a billion people search

World Cup 2026 + Claude + AI clips = $100k a month on autopilot FIFA made YouTube a preferred platform for the tournament, so the search is pouring in, and the real highlights can't keep up > real match footage gets claimed by Content ID and demonetized > AI-generated clips, a packed stadium, flags, a slow-motion goal, belong to no one and keep every dollar > sports CPMs are up 20 to 40% during the tournament > Claude writes the clip, a generator builds it in 3 minutes, no real frame in any of it She's 22 and has never watched a match the edge isn't knowing soccer, it's being on the right side of Content ID while a billion people search

Kiyoro

19,609 views • 2 months ago

A NEW AI VIDEO EDITING TOOL JUST DROPPED THAT WORKS LIKE A REAL VIDEO EDITING TEAM INSTEAD OF A CLIP GENERATOR it's called agent opus and it's not one model doing everything. it's a team of specialized agents handing work to each other like an actual production studio. you type a prompt, attach references (links, PDFs, images), pick a format, and they take it from there: > @ researcher pulls facts from your URLs and docs > @ scriptwriter drafts the transcript > @ storyboardartist breaks it into scenes > @ hookdesigner generates 3 viral hooks for you to pick from > @ assetmanager collects the visuals > @ motiondesigner builds the animations > @ voiceactor handles narration > @ videoeditor cuts the footage and polishes timings you stay in the loop the whole way. it shows you the plan, waits for your sign-off, lets you pick the hook, then assembles the final scenes. prompt box options: vertical format, voice, AI twin, and style match (so it can copy an existing video's look). the bigger picture: the multi-agent pattern that took over coding tools (claude code, codex) just landed in video production. each agent owns one role and only its role. the real read: the bottleneck for AI video was never the clip generation. it was the research, scripting, hooks, and editing decisions. agent opus is the first tool that breaks those into named specialists instead of stuffing them all into one bloated model. link in replies 👇

A NEW AI VIDEO EDITING TOOL JUST DROPPED THAT WORKS LIKE A REAL VIDEO EDITING TEAM INSTEAD OF A CLIP GENERATOR it's called agent opus and it's not one model doing everything. it's a team of specialized agents handing work to each other like an actual production studio. you type a prompt, attach references (links, PDFs, images), pick a format, and they take it from there: > @ researcher pulls facts from your URLs and docs > @ scriptwriter drafts the transcript > @ storyboardartist breaks it into scenes > @ hookdesigner generates 3 viral hooks for you to pick from > @ assetmanager collects the visuals > @ motiondesigner builds the animations > @ voiceactor handles narration > @ videoeditor cuts the footage and polishes timings you stay in the loop the whole way. it shows you the plan, waits for your sign-off, lets you pick the hook, then assembles the final scenes. prompt box options: vertical format, voice, AI twin, and style match (so it can copy an existing video's look). the bigger picture: the multi-agent pattern that took over coding tools (claude code, codex) just landed in video production. each agent owns one role and only its role. the real read: the bottleneck for AI video was never the clip generation. it was the research, scripting, hooks, and editing decisions. agent opus is the first tool that breaks those into named specialists instead of stuffing them all into one bloated model. link in replies 👇

Creators Toolbox

25,785 views • 4 months ago