A.I.Warper's banner

A.I.Warper

@AIWarper • 36,805 subscribers

Creator of the Lil Yachty viral walk out trend

Shorts

Below I will teach you how to reverse engineer any 15s video you see You will need to tweak it a bit but I will explain in this thread exactly how to make these if you are ever curious Thread below 👇

Below I will teach you how to reverse engineer any 15s video you see You will need to tweak it a bit but I will explain in this thread exactly how to make these if you are ever curious Thread below 👇

364,163 Aufrufe

Much easier in 2026 vs 2024 when I made the original Made with Minimax H3

Much easier in 2026 vs 2024 when I made the original Made with Minimax H3

39,905 Aufrufe

GPT-2 + Seedance 2.0 Prompted for my avatar to ask her a question and she replies. If I actually cared about this I would probably manually place the question overlay on the chat overlay.

153,635 Aufrufe

FLUX 3 for me feels like a better SORA. I think people are going to make some really wacky stuff with this model. “Girl from the attached ref images is break dancing on a flattened cardboard box on a bustling NYC sidewalk in 1993”

FLUX 3 for me feels like a better SORA. I think people are going to make some really wacky stuff with this model. “Girl from the attached ref images is break dancing on a flattened cardboard box on a bustling NYC sidewalk in 1993”

50,069 Aufrufe

I wanted to test this previs -> anime workflow I saw posted in the past couple days using Seedance 2 on my own assets to see how well it can do a retro anime vibe. Details below 👇

I wanted to test this previs -> anime workflow I saw posted in the past couple days using Seedance 2 on my own assets to see how well it can do a retro anime vibe. Details below 👇

77,877 Aufrufe

Minimax H3 v2v test This was the first try, I am confident if I worked on the prompt a bit I can get this thing to do 1:1 mocap.

Minimax H3 v2v test This was the first try, I am confident if I worked on the prompt a bit I can get this thing to do 1:1 mocap.

41,308 Aufrufe

Ok this is as close as I can get it.. pretty good one shot replacement. Interested in how I did it? Yay or nay

Ok this is as close as I can get it.. pretty good one shot replacement. Interested in how I did it? Yay or nay

18,344 Aufrufe

T2V with Minimax-H3 locally 960 x 544 resolution generated in 227s

T2V with Minimax-H3 locally 960 x 544 resolution generated in 227s

28,245 Aufrufe

Entirely prompted with Seedance 2 Full process below 👇

Entirely prompted with Seedance 2 Full process below 👇

63,010 Aufrufe

Workflow tutorial: Seedance ref videos to drive your animations Disclaimer: Expect to hit moderation more than usual. Details below in 🧵

Workflow tutorial: Seedance ref videos to drive your animations Disclaimer: Expect to hit moderation more than usual. Details below in 🧵

48,740 Aufrufe

Another viral template idea - The Oscar Win 🏆 Better luck next year PJ Ace jboogx.creative The Dor Brothers 🤣 Process explained below 👇 As always, if this is useful please let me know by engaging 💙

Another viral template idea - The Oscar Win 🏆 Better luck next year PJ Ace jboogx.creative The Dor Brothers 🤣 Process explained below 👇 As always, if this is useful please let me know by engaging 💙

68,325 Aufrufe

New Minimax model handles 2D animation very nice! So far very impressive model.... we are so spoiled these days with options it's a great time to be a creator. Worth noting, the sound on this model is significantly better than some of the other options.. Feels close to SD 2.0

New Minimax model handles 2D animation very nice! So far very impressive model.... we are so spoiled these days with options it's a great time to be a creator. Worth noting, the sound on this model is significantly better than some of the other options.. Feels close to SD 2.0

28,384 Aufrufe

So close.... Google Omni

So close.... Google Omni

63,573 Aufrufe

Step 1: Generate the base image with GPT-2 Step 2: Serve the image you just generated back to GPT-2 with this prompt: "Convert this scene into a 360 equirectangular image"

Step 1: Generate the base image with GPT-2 Step 2: Serve the image you just generated back to GPT-2 with this prompt: "Convert this scene into a 360 equirectangular image"

77,151 Aufrufe

I STRONGLY encourage you to adhere to the prompting guide released by Minimax. It really helps a lot with getting exactly what you expected.... My last post did not follow the suggested prompt structure

I STRONGLY encourage you to adhere to the prompting guide released by Minimax. It really helps a lot with getting exactly what you expected.... My last post did not follow the suggested prompt structure

21,742 Aufrufe

After testing something more difficult, I have deduced that it's in your best interest to segment your videos into clips rather than trying to blast the full shot. This is as close as I could get this ambitious one shot Minimax H3

After testing something more difficult, I have deduced that it's in your best interest to segment your videos into clips rather than trying to blast the full shot. This is as close as I could get this ambitious one shot Minimax H3

12,262 Aufrufe

MINIMAX H3 audio generated from Suno + a single ref image with a prompt for a Cowboy Bepop style intro 🎺

MINIMAX H3 audio generated from Suno + a single ref image with a prompt for a Cowboy Bepop style intro 🎺

21,114 Aufrufe

As I said... give me a min to cook... Full 1 shot, 13 scenes. 1 single reference image. Minimax H3 in ComfyUI with a custom workflow

As I said... give me a min to cook... Full 1 shot, 13 scenes. 1 single reference image. Minimax H3 in ComfyUI with a custom workflow

10,803 Aufrufe

Depth tests with Seedance 2.0 Explanation 👇

Depth tests with Seedance 2.0 Explanation 👇

27,519 Aufrufe

Blender MCP w/ Codex 5.5 or OPUS 4.8 is seriously pretty cracked... This entire block out was one-shot with a single prompt in about 15 minutes. Only problem I am having now is getting Seedance to use it properly 😅 Please feel free to use it and post your results below 👇

Blender MCP w/ Codex 5.5 or OPUS 4.8 is seriously pretty cracked... This entire block out was one-shot with a single prompt in about 15 minutes. Only problem I am having now is getting Seedance to use it properly 😅 Please feel free to use it and post your results below 👇

24,785 Aufrufe

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

This is absolutely insane! 30s video length with very decent stability. Performance on the bottom by JulianoMass on IG How? Foolishly simple 👇

This is absolutely insane! 30s video length with very decent stability. Performance on the bottom by JulianoMass on IG How? Foolishly simple 👇

4,082,626 Aufrufe • vor 7 Monaten

You could build a bespoke influencer right now using only AI. Scrape clips or record yourself. Automate the rest: - scrape - LLM audit - bucket by category - process - voice changer This stack already works - the real unlock is realtime. Clips: Emiru, gabimfmoura

You could build a bespoke influencer right now using only AI. Scrape clips or record yourself. Automate the rest: - scrape - LLM audit - bucket by category - process - voice changer This stack already works - the real unlock is realtime. Clips: Emiru, gabimfmoura

288,104 Aufrufe • vor 7 Monaten

ChatGPT prompted 74 times “Create the exact replica of this image, do not change a thing” This is why I say you need to start a new chat after each edit 🤣

ChatGPT prompted 74 times “Create the exact replica of this image, do not change a thing” This is why I say you need to start a new chat after each edit 🤣

574,460 Aufrufe • vor 1 Jahr

FLUX 3: "a Timelapse video of a kid opening a Lego set and constructing a Star Wars X wing. Full timelapse from box opening to final constructed Lego X wing. Continuous shot with no cuts"

FLUX 3: "a Timelapse video of a kid opening a Lego set and constructing a Star Wars X wing. Full timelapse from box opening to final constructed Lego X wing. Continuous shot with no cuts"

45,683 Aufrufe • vor 1 Monat

Beautiful SD 2.5 output from Keshi. I wanted to see what would happen if I pipped their video through my "reverse engineer" pipeline but with two unique characters I have from a different story. SD 2.5 can make some really great anime.. Prompt and refs in first comment below

Beautiful SD 2.5 output from Keshi. I wanted to see what would happen if I pipped their video through my "reverse engineer" pipeline but with two unique characters I have from a different story. SD 2.5 can make some really great anime.. Prompt and refs in first comment below

16,829 Aufrufe • vor 16 Tagen

Elon dropping major alpha on the latest Rogan podcast It's so insanely simple to make anyone say anything you want these days... these Elon clips especially feels super good. Entire thing took 20 mins....care to know how?

Elon dropping major alpha on the latest Rogan podcast It's so insanely simple to make anyone say anything you want these days... these Elon clips especially feels super good. Entire thing took 20 mins....care to know how?

177,029 Aufrufe • vor 10 Monaten

If you could make this realtime enough such that a Twitch chat could steer the next 15s You'd break the internet... Imagine a real time video model where the chat just decides what happens next haha We all know exactly what would happen

If you could make this realtime enough such that a Twitch chat could steer the next 15s You'd break the internet... Imagine a real time video model where the chat just decides what happens next haha We all know exactly what would happen

19,896 Aufrufe • vor 24 Tagen

"Minimax H3 is trash" Nah bro.... you just haven't a clue 🤣 1 audio file 4 ref images One shot (2x 15s stitched). Non cherry picked results. If there is interest I will explain exactly how I made it 💙 I have a 9:16 variant too I'll share later

"Minimax H3 is trash" Nah bro.... you just haven't a clue 🤣 1 audio file 4 ref images One shot (2x 15s stitched). Non cherry picked results. If there is interest I will explain exactly how I made it 💙 I have a 9:16 variant too I'll share later

23,021 Aufrufe • vor 1 Monat

FLUX 3 also understands image annotations. In hindsight I should have asked it to also remove the red line or not draw it haha Prompt and source images in first comment below

FLUX 3 also understands image annotations. In hindsight I should have asked it to also remove the red line or not draw it haha Prompt and source images in first comment below

20,168 Aufrufe • vor 1 Monat

💙Kendomland - land of the free and the men 🩷 This one was fun and feels unique I left the expressions and mouth movements more muted intentionally as it depicts a "doll" more clearly... I think? As always - please like and share if you enjoy this content! Love you all

💙Kendomland - land of the free and the men 🩷 This one was fun and feels unique I left the expressions and mouth movements more muted intentionally as it depicts a "doll" more clearly... I think? As always - please like and share if you enjoy this content! Love you all

262,363 Aufrufe • vor 2 Jahren

TENCENT casually dropping a 13bn parameter open sourced video model WITH weights 🩵 Looks unreal and can't wait to test it

TENCENT casually dropping a 13bn parameter open sourced video model WITH weights 🩵 Looks unreal and can't wait to test it

156,356 Aufrufe • vor 1 Jahr

My god this is satisfying slop 🤤 From user heeyasound on IG

My god this is satisfying slop 🤤 From user heeyasound on IG

66,875 Aufrufe • vor 7 Monaten

481 frames, 24fps 720 x 1280 resolution Total VRAM: ~40gb Rendered in ComfyUI using LTX 2.3 with audio + image as input. Still playing with settings but again... pretty good out of the box

481 frames, 24fps 720 x 1280 resolution Total VRAM: ~40gb Rendered in ComfyUI using LTX 2.3 with audio + image as input. Still playing with settings but again... pretty good out of the box

55,187 Aufrufe • vor 6 Monaten

Image to depth using Marigold and ConfyUI Then depth to 3D geometry in Blender Super interesting stuff 🤯

Image to depth using Marigold and ConfyUI Then depth to 3D geometry in Blender Super interesting stuff 🤯

201,906 Aufrufe • vor 2 Jahren

OmniVoice Zeroshot Clone + VEED Fabric Test VEED Fabric accepts audio + image and creates a lipsync video. Obviously.... turn your volume up.

OmniVoice Zeroshot Clone + VEED Fabric Test VEED Fabric accepts audio + image and creates a lipsync video. Obviously.... turn your volume up.

18,874 Aufrufe • vor 1 Monat

🚨Release - WarpTuber (A real-time VTubing Tool) (link to download in first comment) 👇 I've included a premade VENV. In a perfect world you just unzip this and it works out of the box (tested on Windows 11) My hopes are that people smarter than me pick it up from here and run with it! More comments below 👇 $WARPER

🚨Release - WarpTuber (A real-time VTubing Tool) (link to download in first comment) 👇 I've included a premade VENV. In a perfect world you just unzip this and it works out of the box (tested on Windows 11) My hopes are that people smarter than me pick it up from here and run with it! More comments below 👇 $WARPER

96,142 Aufrufe • vor 1 Jahr

👨‍🍳The longest continuous shot I have ever done🐻‍❄️ 2 clips: First: 909 frames in 12 context windows. Prompt executed in 00:39:25 Second: 2337 frames in 31 context windows. Prompt executed in 01:32:02 Max allocated VRAM=27.458 GB 720p resolution Need 260gb of System RAM though

👨‍🍳The longest continuous shot I have ever done🐻‍❄️ 2 clips: First: 909 frames in 12 context windows. Prompt executed in 00:39:25 Second: 2337 frames in 31 context windows. Prompt executed in 01:32:02 Max allocated VRAM=27.458 GB 720p resolution Need 260gb of System RAM though

56,900 Aufrufe • vor 10 Monaten

Trained another voice lora using VOXCPM. Also got the TTS running near real time with a Rust port and WSL Absolutely incredible. This one was trained on 20 clips. Not releasing these for obvious reasons - purely for science

Trained another voice lora using VOXCPM. Also got the TTS running near real time with a Rust port and WSL Absolutely incredible. This one was trained on 20 clips. Not releasing these for obvious reasons - purely for science

28,531 Aufrufe • vor 4 Monaten

🐺The "Wolf" of Wall Street 📈 600+ frames on a 5090 Time: 35mins 20 seconds 1024 x 576 res and 4 steps - going to see if I can push the quality higher on a bigger card. Thoughts? The paws are my biggest critique. Believe it's because they exit frame a lot.

🐺The "Wolf" of Wall Street 📈 600+ frames on a 5090 Time: 35mins 20 seconds 1024 x 576 res and 4 steps - going to see if I can push the quality higher on a bigger card. Thoughts? The paws are my biggest critique. Believe it's because they exit frame a lot.

54,227 Aufrufe • vor 11 Monaten

Trained a voice lora on Walken using just 6 clips as a test to see how well VoxCPM handles learning a voice. Then I just did audio + image = video with LTX 2.3. This TTS model feels extremely powerful, fully open sourced (Apache 2.0) More below 👇

Trained a voice lora on Walken using just 6 clips as a test to see how well VoxCPM handles learning a voice. Then I just did audio + image = video with LTX 2.3. This TTS model feels extremely powerful, fully open sourced (Apache 2.0) More below 👇

23,851 Aufrufe • vor 4 Monaten