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.

I love open source AI in ComfyUI because it lets me control every step. For this spaceship shot I tracked the camera, designed layout geometry in Blender, exported depth maps, prompted the final look with an AI image editor, and jumped into Photoshop to clean it all up.

Mickmumpitz

15,092 subscribers

47,986 views • 5 months ago •via X (Twitter)

Arts Education Science & Technology

Anya Rossi• Live Now

Private livecam show

12 Comments

FossyCat5 months ago

@ComfyUI Thank you for providing these amazing tutorials. I've learned so much from your work!

GOD OF POL (previously MATIC)5 months ago

@ComfyUI blender to depth maps to comfy to photoshop, thats a proper pipeline. wonder if youve looked at @SentientAGI's open source agent frameworks, they do something similar but for reasoning in addition to multi tool AI workflows. your VFX approach is basically what they want for AGI

Javi Lopez ⛩️5 months ago

@ComfyUI This is so good!

bdvd_255 months ago

@ComfyUI what video model did you use?, ltx2.3?

Mickmumpitz5 months ago

@ComfyUI

bdvd_255 months ago

@ComfyUI ty, i will look it up

Prince Bell5 months ago

@ComfyUI This is super dope!

Trust⭕️5 months ago

@ComfyUI Bro ur insane! Thats so dope

Unreal Creators Hub5 months ago

@ComfyUI Esta increible!

Grizu5 months ago

@ComfyUI This deserves a follow

Peachy J1 month ago

@ComfyUI Controlling every step instead of trusting the black box has real value.

AJM5 months ago

@ComfyUI What chance do have following along with your workflows with and amd gpu ?

Related Videos

Our newest AI short film was 100% filmed on location. Then using open source video & image models in ComfyUI, we transformed me into a man who wakes up as a monstrous vermin! Complete tutorial + FREE workflows shared below!

Our newest AI short film was 100% filmed on location. Then using open source video & image models in ComfyUI, we transformed me into a man who wakes up as a monstrous vermin! Complete tutorial + FREE workflows shared below!

Mickmumpitz

12,682 views • 1 year ago

This AI video is a Masterpiece 🤯 Took an experiment further than planned: feeding one concept shot into GPT-6 Astra, generating the base geometry in Blender, and pushing the final render through Flova. Made it with FlovaAI

This AI video is a Masterpiece 🤯 Took an experiment further than planned: feeding one concept shot into GPT-6 Astra, generating the base geometry in Blender, and pushing the final render through Flova. Made it with FlovaAI

William AI

606,819 views • 13 days ago

I created a full short film using free open source AI video models. I jumped into the role of the wizard Trogus and transformed myself into him using AI. Imagine what kind of movies you would have made if you had this tool as a kid!

I created a full short film using free open source AI video models. I jumped into the role of the wizard Trogus and transformed myself into him using AI. Imagine what kind of movies you would have made if you had this tool as a kid!

Mickmumpitz

283,455 views • 1 year ago

I built a Photoshop-like image editor. It’s called Compositor. I originally built it for myself to get off my Adobe subscription, but decided to release it for free and make it open source. It has all the essential tools I need for compositing, with none of the BS. I know this workflow is kind of archaic in 2026, but I’m still using it until AI can actually get pixel-perfect on some of the details. The entire app is 12MB. Photoshop is 6,455MB on my machine.

I built a Photoshop-like image editor. It’s called Compositor. I originally built it for myself to get off my Adobe subscription, but decided to release it for free and make it open source. It has all the essential tools I need for compositing, with none of the BS. I know this workflow is kind of archaic in 2026, but I’m still using it until AI can actually get pixel-perfect on some of the details. The entire app is 12MB. Photoshop is 6,455MB on my machine.

Robbie Tilton

2,150,551 views • 5 days ago

Testing out VirtuCamera for Blender 🔶 with one of our HD head scans. Super easy to use and gives you a really nice handheld camera feel. I did the camera work in Blender, then just exported the camera as an FBX to Marmoset and rendered it all in about an hour.

Testing out VirtuCamera for Blender 🔶 with one of our HD head scans. Super easy to use and gives you a really nice handheld camera feel. I did the camera work in Blender, then just exported the camera as an FBX to Marmoset and rendered it all in about an hour.

3D Scanstore

87,827 views • 2 years ago

Adobe just dropped Assisted Editor (beta) in Photoshop and it's wild 🤯 You can literally drag a slider to change the lighting of an image with AI I’ve been using it to create frames for my AI videos! Super useful. It's in main Photoshop (not beta).

Adobe just dropped Assisted Editor (beta) in Photoshop and it's wild 🤯 You can literally drag a slider to change the lighting of an image with AI I’ve been using it to create frames for my AI videos! Super useful. It's in main Photoshop (not beta).

Kris Kashtanova

12,333 views • 26 days ago

Okay this shouldn't work. Krea 2 is an image generator, not an edit model. But thanks to the amazing work by the amazing ComfyUI community I was able to use it for my Consistent Character Creator and it works super well!

Okay this shouldn't work. Krea 2 is an image generator, not an edit model. But thanks to the amazing work by the amazing ComfyUI community I was able to use it for my Consistent Character Creator and it works super well!

Mickmumpitz

17,191 views • 1 month ago

✨ I tried Claude Code cause Marc Köhlbrugge kept telling me too and asked it to make this inpaint editor for Photo AI I had this idea of some 🌈 holographic paint brush strokes so I just asked it and it one shotted this glowing JS code for , extremely cool because I could never make this myself without AI All vanilla jQuery code also no libraries (as usual) The inpaint editor lets you edit any of your phtoos AND it uses your AI model so even if you touch the face it still looks like your model unlike regular AI image editors! Also this brings my editing (and all my features now) in-house (before it was an embedded Photopea editor)

✨ I tried Claude Code cause Marc Köhlbrugge kept telling me too and asked it to make this inpaint editor for Photo AI I had this idea of some 🌈 holographic paint brush strokes so I just asked it and it one shotted this glowing JS code for , extremely cool because I could never make this myself without AI All vanilla jQuery code also no libraries (as usual) The inpaint editor lets you edit any of your phtoos AND it uses your AI model so even if you touch the face it still looks like your model unlike regular AI image editors! Also this brings my editing (and all my features now) in-house (before it was an embedded Photopea editor)

@levelsio

256,917 views • 1 year ago

I used AI to turn a random image into this 3,847-piece dragon boss 🐉 everyone's sleeping on how good Tripo AI has gotten for 3D modeling, so I put it to the test. uploaded one reference image → generated the helmet, the armor, every scale piece by piece → brought it into Cinema4D for the final layout. that moment when the eyes light up red? that wasn't hand-animated frame by frame. that was AI-generated geometry doing the heavy lifting while I focused on the story. the 3D workflow just got way more interesting. full breakdown 👇

I used AI to turn a random image into this 3,847-piece dragon boss 🐉 everyone's sleeping on how good Tripo AI has gotten for 3D modeling, so I put it to the test. uploaded one reference image → generated the helmet, the armor, every scale piece by piece → brought it into Cinema4D for the final layout. that moment when the eyes light up red? that wasn't hand-animated frame by frame. that was AI-generated geometry doing the heavy lifting while I focused on the story. the 3D workflow just got way more interesting. full breakdown 👇

Darshal Jaitwar

252,900 views • 14 days ago

Photoshop AI Agent / MCP example (using Claude), where I couple it with an image vision MCP to have Claude AI rename the layers in a Photoshop file based on the content of the layers. Thanks to Paul Trani for supplying a Photoshop file with a typical naming scheme (none).

Photoshop AI Agent / MCP example (using Claude), where I couple it with an image vision MCP to have Claude AI rename the layers in a Photoshop file based on the content of the layers. Thanks to Paul Trani for supplying a Photoshop file with a typical naming scheme (none).

Mike Chambers

16,800 views • 1 year ago

I made another film with Ai. This time I used #runwayml ‘s text to video Ai #gen2 . Each shot was prompted. Voices generated with #elevenlabs I wrote the story based on a tweet #gpt4 wrote: “I just had a dream that I was an AI and woke up in a lab. What does it mean?” #aicinema

I made another film with Ai. This time I used #runwayml ‘s text to video Ai #gen2 . Each shot was prompted. Voices generated with #elevenlabs I wrote the story based on a tweet #gpt4 wrote: “I just had a dream that I was an AI and woke up in a lab. What does it mean?” #aicinema

Jon Finger

270,349 views • 3 years ago

Claude Fable 5 has been out for a day. I already one-shotted a voxel spaceship racer with it. I fed it a reference image of a track layout and told it to build a full game around that shape. Missiles, boost zones, 3 AI opponents, 3 laps, a HUD with speed and health. Fable 5 read the image, traced the track geometry, and built the entire thing. One prompt. This is what Mythos-class models do to the build cycle.

Claude Fable 5 has been out for a day. I already one-shotted a voxel spaceship racer with it. I fed it a reference image of a track layout and told it to build a full game around that shape. Missiles, boost zones, 3 AI opponents, 3 laps, a HUD with speed and health. Fable 5 read the image, traced the track geometry, and built the entire thing. One prompt. This is what Mythos-class models do to the build cycle.

God of Prompt

27,386 views • 3 months ago

AI just fired the video editor 3D Motion just dropped the fastest way to turn any image into a cinematic video. With depth mapping, smooth camera moves, and movie-like lighting, it makes static photos look like Hollywood shots. Tutorial below + Prompts:

AI just fired the video editor 3D Motion just dropped the fastest way to turn any image into a cinematic video. With depth mapping, smooth camera moves, and movie-like lighting, it makes static photos look like Hollywood shots. Tutorial below + Prompts:

Robin Delta

137,796 views • 1 year ago

AI is taking over Hollywood.. I just made a VHS music video with Higgsfield AI in 2 hours, film studio level camera work, natural human body movement this is the first AI short I made that doesn't look like AI at all step by step tutorial with prompts for each shot:

AI is taking over Hollywood.. I just made a VHS music video with Higgsfield AI in 2 hours, film studio level camera work, natural human body movement this is the first AI short I made that doesn't look like AI at all step by step tutorial with prompts for each shot:

el.cine

209,868 views • 1 year ago

I just automated the entire process of creating hyperrealistic consistent AI character datasets. Completely free on your own hardware. Everything runs locally using free, open-source models in ComfyUI. FULL TUTORIAL & FREE WORKFLOWS👇

I just automated the entire process of creating hyperrealistic consistent AI character datasets. Completely free on your own hardware. Everything runs locally using free, open-source models in ComfyUI. FULL TUTORIAL & FREE WORKFLOWS👇

Mickmumpitz

32,115 views • 11 months ago

Made a simple and cute Channel Packer as a weekend project :) Means I don't have to open Photoshop every time I want to change the channels or the textures and at the same time lets me preserve my source textures (3 channel normal maps, b&w non-packed masks...). So happy with it!

Made a simple and cute Channel Packer as a weekend project :) Means I don't have to open Photoshop every time I want to change the channels or the textures and at the same time lets me preserve my source textures (3 channel normal maps, b&w non-packed masks...). So happy with it!

Alex

23,540 views • 3 years ago

Paid partnership with NVIDIA Studio - You can now render entire CG movies with local AI in ComfyUI on your own computer, for free. I built the whole pipeline with LTX.io and used it to make a short film about a motel run by bears. 🐻 Free workflows + full tutorial 👇

Paid partnership with NVIDIA Studio - You can now render entire CG movies with local AI in ComfyUI on your own computer, for free. I built the whole pipeline with LTX.io and used it to make a short film about a motel run by bears. 🐻 Free workflows + full tutorial 👇

Mickmumpitz

35,971 views • 3 months ago

GUYS I MADE IT AND IT WORKS :OO Asked ChatGPT-4o and it couldn't do it Then switched to o1 and it did it after 2 tries Now all I need to do is move the camera, take the image + blackmask, ask AI to fill in the black part, then de-sphere it and we have a 360' camera control!!!

GUYS I MADE IT AND IT WORKS :OO Asked ChatGPT-4o and it couldn't do it Then switched to o1 and it did it after 2 tries Now all I need to do is move the camera, take the image + blackmask, ask AI to fill in the black part, then de-sphere it and we have a 360' camera control!!!

@levelsio

271,661 views • 1 year ago