Cool prompt. Made it 3D with #CogVideoX
March 1, 2025 at 12:05 PM
Cool prompt. Made it 3D with #CogVideoX
1/2
"One five-second AI video burns what a microwave uses in an hour.
MIT Technology Review did the math on an open-source AI video tool called CogVideoX. A single five-second clip took about 3.4 million joules of energy."
Sources: MIT Technology Review, International Energy Agency
"One five-second AI video burns what a microwave uses in an hour.
MIT Technology Review did the math on an open-source AI video tool called CogVideoX. A single five-second clip took about 3.4 million joules of energy."
Sources: MIT Technology Review, International Energy Agency
September 24, 2026 at 12:58 PM
1/2
"One five-second AI video burns what a microwave uses in an hour.
MIT Technology Review did the math on an open-source AI video tool called CogVideoX. A single five-second clip took about 3.4 million joules of energy."
Sources: MIT Technology Review, International Energy Agency
"One five-second AI video burns what a microwave uses in an hour.
MIT Technology Review did the math on an open-source AI video tool called CogVideoX. A single five-second clip took about 3.4 million joules of energy."
Sources: MIT Technology Review, International Energy Agency
VEO3 looks like it's ~30fps
CogVideoX 8fps took 100k joules / 5 seconds
CogVideoX 16fps took 3.4m joules / 5 seconds
Double the length, double the fps, assuming efficiency gains counteract the exponential curve here but that's a big assumption
CogVideoX 8fps took 100k joules / 5 seconds
CogVideoX 16fps took 3.4m joules / 5 seconds
Double the length, double the fps, assuming efficiency gains counteract the exponential curve here but that's a big assumption
May 21, 2025 at 2:58 PM
VEO3 looks like it's ~30fps
CogVideoX 8fps took 100k joules / 5 seconds
CogVideoX 16fps took 3.4m joules / 5 seconds
Double the length, double the fps, assuming efficiency gains counteract the exponential curve here but that's a big assumption
CogVideoX 8fps took 100k joules / 5 seconds
CogVideoX 16fps took 3.4m joules / 5 seconds
Double the length, double the fps, assuming efficiency gains counteract the exponential curve here but that's a big assumption
Experiments with #CogVideoX Um, scary #Loomered
January 5, 2025 at 3:41 PM
Experiments with #CogVideoX Um, scary #Loomered
Wow, beautiful! 😍 I didn’t know but will check it out! Thank you for the welcome, video, and tip. 💖
#CogVideoX
#Aiart #midjourney #synthart
#CogVideoX
#Aiart #midjourney #synthart
Did you know you can make your creations 3D with #CogVideoX ? Pretty fun. Welcome to the community
March 1, 2025 at 3:52 PM
Wow, beautiful! 😍 I didn’t know but will check it out! Thank you for the welcome, video, and tip. 💖
#CogVideoX
#Aiart #midjourney #synthart
#CogVideoX
#Aiart #midjourney #synthart
Tested @netflix.com VOID AI cleanup model
medium.com/p/netflix-vo...
Interesting POC, but not ready for production VFX yet.
#VFX #AI #VideoInpainting #NetflixVOID #ComfyUI #PostProduction #CogVideoX #VFXPipeline #GenerativeAI #FilmTech #CompositingAI #AIForFilm #VideoCleanup #MachineLearning
medium.com/p/netflix-vo...
Interesting POC, but not ready for production VFX yet.
#VFX #AI #VideoInpainting #NetflixVOID #ComfyUI #PostProduction #CogVideoX #VFXPipeline #GenerativeAI #FilmTech #CompositingAI #AIForFilm #VideoCleanup #MachineLearning
Netflix VOID, part 1: first contact
Tested Netflix VOID — a new open-source video cleanup AI model built on CogVideoX. Released April 2026 under Apache 2.0.
medium.com
May 24, 2026 at 5:00 AM
Tested @netflix.com VOID AI cleanup model
medium.com/p/netflix-vo...
Interesting POC, but not ready for production VFX yet.
#VFX #AI #VideoInpainting #NetflixVOID #ComfyUI #PostProduction #CogVideoX #VFXPipeline #GenerativeAI #FilmTech #CompositingAI #AIForFilm #VideoCleanup #MachineLearning
medium.com/p/netflix-vo...
Interesting POC, but not ready for production VFX yet.
#VFX #AI #VideoInpainting #NetflixVOID #ComfyUI #PostProduction #CogVideoX #VFXPipeline #GenerativeAI #FilmTech #CompositingAI #AIForFilm #VideoCleanup #MachineLearning
November 13, 2024 at 8:30 PM
Reconstruct, Inpaint, Finetune: Dynamic Novel-view Synthesis from Monocular Videos
Kaihua Chen, @tarashakhurana.bsky.social, Deva Ramanan
tl;dr: in title; fine-tune CogVideoX->train 2D video-inpainter
arxiv.org/abs/2507.12646
Kaihua Chen, @tarashakhurana.bsky.social, Deva Ramanan
tl;dr: in title; fine-tune CogVideoX->train 2D video-inpainter
arxiv.org/abs/2507.12646
July 18, 2025 at 10:22 AM
Reconstruct, Inpaint, Finetune: Dynamic Novel-view Synthesis from Monocular Videos
Kaihua Chen, @tarashakhurana.bsky.social, Deva Ramanan
tl;dr: in title; fine-tune CogVideoX->train 2D video-inpainter
arxiv.org/abs/2507.12646
Kaihua Chen, @tarashakhurana.bsky.social, Deva Ramanan
tl;dr: in title; fine-tune CogVideoX->train 2D video-inpainter
arxiv.org/abs/2507.12646
I ran this through #CogVideoX to make it 3D
December 31, 2024 at 9:58 PM
I ran this through #CogVideoX to make it 3D
Some open source video tool alternatives (from huggingface):
CogVideoX
Mochi 1
LTX Video
Pyramid Flow
CogVideoX
Mochi 1
LTX Video
Pyramid Flow
November 27, 2024 at 3:36 PM
Some open source video tool alternatives (from huggingface):
CogVideoX
Mochi 1
LTX Video
Pyramid Flow
CogVideoX
Mochi 1
LTX Video
Pyramid Flow
There's a new AI model that takes images and makes them 3D #CogVideoX
January 1, 2025 at 2:46 PM
There's a new AI model that takes images and makes them 3D #CogVideoX
AP Workflow 12.0 Early Access 2 is out! It comes with support for SD3.5 Large, CogVideoX for T2V and I2V video generation, and a trajectory editor for videos powered by the new Alibaba Tora model.
#SD35 #ComfyUI #CogVideoX #APW
#SD35 #ComfyUI #CogVideoX #APW
October 23, 2024 at 8:56 AM
AP Workflow 12.0 Early Access 2 is out! It comes with support for SD3.5 Large, CogVideoX for T2V and I2V video generation, and a trajectory editor for videos powered by the new Alibaba Tora model.
#SD35 #ComfyUI #CogVideoX #APW
#SD35 #ComfyUI #CogVideoX #APW
Trying Cogvideox in local, might post less in future days due to extensive power needed.
#furryart #furry #furryartist #furryartwork #furryartnsfw #oc #AIart #furai #aifurry #furryai #ai #AI_illustration #AIArtCommuity #AIArtworks18 #Aiartoworks #yiff #porn #aiyiff #aiporn #nsfw
#furryart #furry #furryartist #furryartwork #furryartnsfw #oc #AIart #furai #aifurry #furryai #ai #AI_illustration #AIArtCommuity #AIArtworks18 #Aiartoworks #yiff #porn #aiyiff #aiporn #nsfw
November 1, 2024 at 8:12 PM
Trying Cogvideox in local, might post less in future days due to extensive power needed.
#furryart #furry #furryartist #furryartwork #furryartnsfw #oc #AIart #furai #aifurry #furryai #ai #AI_illustration #AIArtCommuity #AIArtworks18 #Aiartoworks #yiff #porn #aiyiff #aiporn #nsfw
#furryart #furry #furryartist #furryartwork #furryartnsfw #oc #AIart #furai #aifurry #furryai #ai #AI_illustration #AIArtCommuity #AIArtworks18 #Aiartoworks #yiff #porn #aiyiff #aiporn #nsfw
New cute ginger cat made with #AI 🐱
The image was generated using FLUX Pro
The video was generated using CogVideoX-5B 👉 huggingface.co/spaces/THUDM...
The image was generated using FLUX Pro
The video was generated using CogVideoX-5B 👉 huggingface.co/spaces/THUDM...
November 21, 2024 at 4:44 PM
New cute ginger cat made with #AI 🐱
The image was generated using FLUX Pro
The video was generated using CogVideoX-5B 👉 huggingface.co/spaces/THUDM...
The image was generated using FLUX Pro
The video was generated using CogVideoX-5B 👉 huggingface.co/spaces/THUDM...
Okay my reading comprehension is crap because I looked again and I'm pretty sure it's about Ziphu AI's CogVideoX
May 25, 2025 at 1:43 AM
Okay my reading comprehension is crap because I looked again and I'm pretty sure it's about Ziphu AI's CogVideoX
Continuing the @netflix.com VOID test from yesterday.
medium.com/p/netflix-vo...
#VFX #AI #VideoInpainting #NetflixVOID #ComfyUI #PostProduction #OpenSource #CogVideoX #VFXPipeline #GenerativeAI #FilmTech #CompositingAI #AIForFilm #VideoCleanup #MachineLearning #AppleSilicon #M4Max #ACES
medium.com/p/netflix-vo...
#VFX #AI #VideoInpainting #NetflixVOID #ComfyUI #PostProduction #OpenSource #CogVideoX #VFXPipeline #GenerativeAI #FilmTech #CompositingAI #AIForFilm #VideoCleanup #MachineLearning #AppleSilicon #M4Max #ACES
Netflix VOID, part 2: a Mac Studio expedition
Continuing the VOID test from yesterday. This time the goal was specific: can the model slot into a real cinema VFX pipeline on Apple…
medium.com
May 24, 2026 at 5:05 AM
Continuing the @netflix.com VOID test from yesterday.
medium.com/p/netflix-vo...
#VFX #AI #VideoInpainting #NetflixVOID #ComfyUI #PostProduction #OpenSource #CogVideoX #VFXPipeline #GenerativeAI #FilmTech #CompositingAI #AIForFilm #VideoCleanup #MachineLearning #AppleSilicon #M4Max #ACES
medium.com/p/netflix-vo...
#VFX #AI #VideoInpainting #NetflixVOID #ComfyUI #PostProduction #OpenSource #CogVideoX #VFXPipeline #GenerativeAI #FilmTech #CompositingAI #AIForFilm #VideoCleanup #MachineLearning #AppleSilicon #M4Max #ACES
🔥 Hot Repo！ 🔥 (100+ new stars)
📦 THUDM / CogVideo
⭐ 6,169 (+112)
🗒 Python
Text-to-video generation: CogVideoX (2024) and CogVideo (ICLR 2023)
📦 THUDM / CogVideo
⭐ 6,169 (+112)
🗒 Python
Text-to-video generation: CogVideoX (2024) and CogVideo (ICLR 2023)
GitHub - THUDM/CogVideo: Text-to-video generation: CogVideoX (2024) and CogVideo (ICLR 2023)
Text-to-video generation: CogVideoX (2024) and CogVideo (ICLR 2023) - THUDM/CogVideo
github.com
August 29, 2024 at 1:00 PM
A third change in the upcoming EA4 pre-release: APW now supports LoRAs for CogVideoX videos!
The source image (some of you are familiar with it) is 1344x768px generated with FLUX 1 Dev. The video is conditioned with DimensionX Orbit Left LoRA.
#ComfyUI #CogVideoX #APW
The source image (some of you are familiar with it) is 1344x768px generated with FLUX 1 Dev. The video is conditioned with DimensionX Orbit Left LoRA.
#ComfyUI #CogVideoX #APW
November 8, 2024 at 4:42 PM
A third change in the upcoming EA4 pre-release: APW now supports LoRAs for CogVideoX videos!
The source image (some of you are familiar with it) is 1344x768px generated with FLUX 1 Dev. The video is conditioned with DimensionX Orbit Left LoRA.
#ComfyUI #CogVideoX #APW
The source image (some of you are familiar with it) is 1344x768px generated with FLUX 1 Dev. The video is conditioned with DimensionX Orbit Left LoRA.
#ComfyUI #CogVideoX #APW