2024 was a WILD Year for AI Video! Check out HubSpot's Free 1000 Prompts here: https://clickhubspot.com/fy4o
What a wild ride 2024 has been for creative AI! From groundbreaking launches to unexpected twists, this year was nothing short of incredible. In this video, I take you month by month through the biggest developments in AI technology. Whether it’s MidJourney’s evolution, OpenAI’s Sora, or the game-changing tools like Runway Gen-3 and Stable Diffusion 3, we’ve got it all covered. Don’t miss out on how AI shaped the creative world this year!
Thanks to today's sponsor, Hubspot! Download the 1000 Marketing and Productivity Prompts here: https://clickhubspot.com/fy4o
Chapters
0:00 – Intro: A Year of Creative AI
0:29 – January: Multi-Motion Brushes with Runway
0:55 – February: Sora’s Iconic Debut
2:24 – March: AI Music and Talking Heads Revolution
3:51 - Hubspot's 1000 Prompts
5:18 – April: Firefly Video and Dream Machine
7:00 – May: ChatGPT Voice Assistant Controversy
7:52 – June: Gen-3 and Dream Machine Go Head-to-Head
9:04 – July: Talking Heads and Quiet Progress
9:30 – August: Black Forest Labs Flux Dominates
10:29 – September: MiniMax Steals the Spotlight
11:10– October: Meta’s MovieGen and Spooky Updates
11:49 – November: Training Models Everywhere
12:07 – December: Sora’s Disappointing Release and the Future
13:39 – Outro: Thanks for an Incredible 2024
Let's dive into Recraft, the AI image generation platform making waves with its unique model! In this in-depth review, I'll explore everything Recraft has to offer creators and designers.
From generating stunning visuals with strong prompt adherence to advanced features like color adjustments, area modification, vector editing, and powerful image sets, I've got you covered!
Key features covered:
Red Panda (Recraft Version 3) Model
Prompt Adherence
Diverse Artistic Styles
Color Adjustments & Palettes
Area Modification (Lasso Tool)
Image Variations & Aspect Ratios
Vector Editing (SVG Export)
Image Sets (Storyboarding)
Creating Custom Styles
Check out Recraft and use code MEDIA12 for a $12 discount on all plans – https://go.recraft.ai/theoreticallymedia
0:00 Introduction to Recraft
0:33 Who Is Recraft For?
1:08 What was the Red Panda Model?
1:55 Creating Your First Image
2:57 Initial Results & Basic Adjustments
3:45 Advanced Editing: Adding Elements and Modifying Areas
4:28 Creating Image Variations and Different Aspect Ratios
5:21 Exploring Diverse Artistic Styles
6:27 Color Palettes and Style Customization
7:35 Vector Editing and SVG Export
8:30 Palette Options and Storyboarding
8:50 Using Image Sets for Storytelling
9:31 Mini Short Film
9:43 - Image Sets
10:27 Refining Image Sets and Adding Text
11:51 Creating Your Own Style
12:25 Final Thoughts
Google's Veo 2 Is Stunning! Check out HubSpot's Free ChatGPT resources here: https://clickhubspot.com/wwow
In this video, I dive into Google’s newly released Veo 2 AI video generation model, just one week after the arrival of another major player in the AI video scene.
How does Veo-2 measure up, and is it really the new king of AI video? I put it to the test with a series of prompts—from photorealistic island crash landings to eerie ’80s horror vibes, gritty sci-fi settings, and beyond. I’ll share insights into the UI, show off early outputs, and offer tips on prompting for better results. As this is an early-access look, the model still has quirks and limitations, but the leaps in video realism, character physics, and scene composition are undeniably impressive.
LINKS:
Google Veo-2 Waitlist: https://labs.google/experiments
Google Labs Discord: https://discord.gg/googlelabs
Hubspot: https://clickhubspot.com/wwow
Chapters:
0:00 - Intro
0:25 - Setting the Stage
1:09 - Veo Interface
1:28 - First Run
2:16 - Multiple Outputs
2:25 - Video Alternates
3:00 - Abstract Prompts
3:18 - Basic Prompts - 80s Horror Movie
3:44 - Sci-Fi Prompts and Movement
4:27 - Character Design
5:03 - Hubspot ChatGPT resources
6:19 - Veo 2 Prompt Formula
7:00 - Prompt Formula Results
7:27 Other Findings
7:49 Parkour Movments
8:37 - Fight Scenes
9:27 - Image to Video
10:07 - Learning Curve
10:45 - What we Normally Get
11:49 Veo-2 Vs Sora
12:23 Features I'd like to see
13:00 Tips to get access
OpenAI's Sora has a super power and today we're going to dive into it!
In this in-depth exploration, we test Sora’s remarkable ability to remix, upscale, and transform videos—turning simple footage or AI-generated clips into cinematic, photorealistic experiences. We’ll dive into prompt engineering, explore the Recut feature, and see how Sora pairs with other tools like Runway a Is it worth the $200/month price tag? You decide, as I reveal insider tips, show real examples, and discuss how to strategically maximize your subscription for next-level content creation.
My thanks to LTX Studios for sponsoring this video!
Please check them out here: https://bit.ly/LTXVtheo
CHAPTERS
00:00 – Introduction: Sora’s Hidden Superpower
00:32 – Understanding Sora’s Turbo Model vs. “Big Daddy” Version
01:18 – Initial Disappointments & Public Expectations
02:02 – Discovering the True Superpower: Video-to-Video Transformation
02:25 – Remixing Videos: From Midjourney & Minimax to Sora
03:24- Kling 1.0 to Sora
03:51 - Treadmill Girl Example
04:24 - Remix Strength Controls
05:10 - Using Your Own Footage
05:53 - Trained Models and Face Swapping
06:27 - Is It Perfect? It Is Not
07:14 - But it is Still Better Than Image To Video
07:30 - Now Lets Get Weird
07:57 - Using Recut Now
08:34 - Now We Have A Fantasy Show
09:00 - Shoutout To LTX Studio
11:05 - Let's Bash Runway's Act-1
11:26 - Revisiting Tuesday
11:38 - First Fails
12:00 - Solving Consistency Problems
12:28 - Bringing it to Runway
12:54 - Tuesday The Sora Version
13:16 - The Current State of Sora
13:40 - Is it Worth $200 a Month?
Sora has officially launched, and I’m diving into this dramatic AI video generator with its unique features like blending, storyboarding, and remixing. In this video, I explore the tool’s capabilities, test its outputs, and share whether it’s living up to its hype. Plus, I answer the big question: Is it worth the hefty $200 monthly subscription? Stick around to see how Sora stacks up against the competition!
00:00 - Introduction: The Saga of Sora
00:26 - First Impressions of the Sora Interface
01:08 - Sora Website
01:26 - First Test: Man in a Blue Suit
02:03 - You Aren't Going to Like This
02:34 - Image-to-Video Results
03:17 - Image Referencing?
03:33 - Exploring the Storyboard Feature
04:21 - Testing Blend & Remix Features
05:33 - Video Recut in Sora
06:07 - More Image To Video Results
07:11- But The Pirate Girl Works?
07:39 - MKBHD can get Image to Video?
08:00 - Blend Results
08:32 - Recut Results
08:58 - Remix Results
09:14 - More Image-to-Video Experiments
09:54 - Testing Copyright Content in Sora
10:13 - More Image to Video Tests
10:45 - 80s Action Movie Image Prompt
11:29 - Is Sora All That?
12:07 - Is it Worth $200?
Discover the latest AI image generation breakthrough with Leonardo’s brand-new “Flow State” workflow! In this video, we explore how Flow State dramatically streamlines the prompting process, allowing you to generate stunning variations with ease. If you’ve been wrestling with complex prompts, keyword spam, and endless trial-and-error, this feature could be your game-changer. We’ll also revisit some of Leonardo’s powerful legacy tools, discuss their free plan updates, and see how you can leverage Flow State for concept creation, mood boards, and fresh aesthetics. Ready to elevate your AI art journey? Let’s dive right in!
What You’ll Learn:
• How Flow State simplifies AI image generation and reduces prompt complexity
• Tips on exploring multiple aesthetic styles, lighting, and shot types
• How to use the built-in upscaler for cleaner, more detailed results
• The benefits of mixing legacy modes with the new Flow State
• Ideas for concept art, mood boards, and rapid style exploration
Helpful Links & Resources:
Leonardo AI: https://bit.ly/4eKioGw
Chapters:
00:00 – Introduction: Did Leonardo Just Kill Prompting?
00:20 – New “Flow State” Feature Overview
00:30 – Understanding the “Why” Behind Flow State
01:16 – Natural Language Prompts vs. Old Prompting Styles
01:25 – Exploring Styles, Lighting, & Shot Types with Flow State
01:41 – Flow State in Action: Generating Rapid Variations
02:30 – Adjusting Aesthetics & Experimenting with Themes
03:12 – Handling Imperfections & AI Incoherence
04:04 – Using Flow State for Concept Art & Mood Boards
04:43 - Short Video Example
05:04 – Encouraging Creative Exploration & New Aesthetics
05:15 – Trying Prompts You’d Never Use Traditionally
05:55 - Using Flow State to Explore Standard Prompts
06:16 – Diving Deeper: “More Like This” Feature & Style Reinforcement
06:59 – Revisiting Legacy Models & Old-Style Prompting
07:38 – Upscaling Images with the New & Legacy Upscalers
08:24 – Style Referencing & Medium Strength vs. Max Strength
08:47 – Mixed Media Outputs & Unique Combinations
09:16 – Using the Canvas Feature for Inpainting & Refinements
09:47 - Other Features in Canvas
10:23 – Upcoming Features & Real-Time Canvas Fun
10:48 – Plans for 2025 & Leonardo’s Ongoing Free Plan
What if you could generate an open-world video game just from a single prompt? 🚀 Google DeepMind’s Genie 2 takes a groundbreaking leap forward in AI-driven world modeling, bringing us closer to that reality. In this video, I break down everything you need to know about Genie 2: what it is, how it works, and how it stacks up to previous innovations like Genie 1 and similar projects. Plus, we’ll dive into its quirks, limitations, and what this means for the future of AI in gaming. Spoiler: it’s not quite ready for prime time, but the potential is undeniable!
Previous Video on AI Doom: https://youtu.be/jHS1RJREG2Q
Video on World Labs (Second Half of the Video): https://youtu.be/gCZCTor2Xco
Chapters
00:00 - Intro
00:25 - What is Genie 2?
00:55 - Advancement over Genie 1
01:18 - How Genie 2 Works: World Models Explained
01:40 - The Big Breakthough with Genie 2
01:52 - What is a World Model
02:35 - Generating What It Has Never Seen
02:55 - Persistent Memory With Genie 2
03:24 - Limitations
4:00 - Softness in the Output
04:38 - 3d Understanding
04:57 - More Examples of Object Interaction
05:19 - Character Animation in Genie 2
05:31 - NPCs & AI Interactivity: Sima Integration
06:20 - AI Bloopers and Fun Outtakes
07:06 - Competing Projects & What’s Next
Today I'm diving into the future of AI-generated video and imagery! From Tencent, the groundbreaking open-source Hunyuan Video model to World Labs, which lets you generate 3D worlds from a single image, this video is packed with exciting updates. Plus, I’ll share breaking news about Sora, Korea’s new image editor, and image-to-video advancements that are shaking up the creative AI space. Don’t miss this deep dive into the cutting-edge tech shaping 2025!
Links to topics covered:
Hunyuan Video Model: https://github.com/Tencent/HunyuanVideo/tree/main
PadPhone Spot: https://x.com/lepadphone/status/1863924360526094540
WorldLabs: https://www.worldlabs.ai/blog
World Labs Waitlist: https://docs.google.com/forms/d/e/1FAIpQLSf9jHsaDq1IwM_FADQP0Gbd82tbW4CBOI5YfUAdPfqrFrWEeA/viewform
Runway for AI Video Effects: https://runwayml.com/
Krea' s Image Editor Beta: https://www.krea.ai/home
Minimax Image-to-Video: https://hailuoai.video/
00:00 - Intro: The future of AI-generated media
00:23 - Hunyuan Video Model: 13B parameters, open-source powerhouse
01:14 - Example Outputs: SWAT guy, motorcycles, and gym thirst traps
02:52 - Multi Camera Shots
03:33 - Concept Generalization
03:50 - Breakthrough Technology: Dual-stream model design
04:34 - Padfone Ad Spot: Stunning visuals with text-to-video
05:20 - Generated Sound: Birds chirping and spooky footsteps
06:30 - Lip Sync & Gestures: New possibilities for storytelling
07:02 - Single Image to TikTok Dances: From anime girls to statues
07:24 - Expression Control
07:45 - System Requirements
08:21 - World Labs: Explore 3D worlds from one image
09:22 - Difference from Tencent's GameGen AI
09:43 - How World Labs is really impressive
10:28 - Effects in WorldLabs
11:40 - Runway & Generated Environments: Retro VHS vibes
12:24 - Breaking News: Korea’s image editor, Minimax, and Sora updates
Let me know in the comments: Are you excited about open-source video models or World Labs? Or are you just here for the gym treadmill thirst traps?
Open AI's Sora Turbo Model was recently leaked for a handful of hours on Hugging Face. This is (to my knowledge) every user generated video and their prompts.
This leak was orchestrated by a group of artists who were granted beta-access to Sora, who released a statement which you can find below.
The videos presented here represent most of what was generated by Sora on Hugging Face through use of an API. It is likely the "optimized" Turbo Mode of Sora, and it is unclear if this is the latest version.
I normally like to credit the original generators, but in this case, that proved it be a bit impossible. So, all credit to them. I tried, but never got a generation off. Sigh.
As a note, i did lay down some ambient music I preciously created (Not AI generated!) just to have some kind of audio running.
The bulk of the generations were sourced from DesuAnon. If you want to download the mp4 source files, you can do so here:
https://huggingface.co/desuAnon/SoraVids
If you want to scroll through quickly, here are chapter breaks:
00:00 - Sora Video 1
00:10 - Sora Video 2
00:20 - Sora Video 3
00:30 - Sora Video 4
00:40 - Sora Video 5
00:50 - Sora Video 6
01:00 - Sora Video 7
01:10 - Sora Video 8
01:15 - Sora Video 9
01:25 - Sora Video 10
01:35 - Sora Video 11
01:45 - Sora Video 12
01:55 - Sora Video 13
02:05 - Sora Video 14
02:15 - Sora Video 15
02:25 - Sora Video 16
02:35 - Sora Video 17
02:45 - Sora Video 18
02:55 - Sora Video 19
03:05 - Sora Video 20
03:15 - Sora Video 21
03:25 - Sora Video 22
03:35 - Sora Video 23
03:45 - Sora Video 24
03:55 - Sora Video 25
04:05 - Sora Video 26
04:15 - Sora Video 27
04:25 - Sora Video 28
04:35 - Sora Video 29
04:45 - Sora Video 30
04:55 - Sora Video 31
05:05 - Sora Video 32
05:10 - Sora Video 33
05:20 - Sora Video 34
05:25 - Sora Video 35
05:35 - Sora Video 36
05:45 - Sora Video 37
05:55 - Sora Video 38
06:05 - Sora Video 39
06:15 - Sora Video 40
06:25 - Sora Video 41
06:35 - Sora Video 42
06:45 - Sora Video 43
06:55 - Sora Video 44
07:05 - Sora Video 45
07:15 - Sora Video 46
07:25 - Sora Video 47
07:35 - Sora Video 48
07:45 - Sora Video 49
07:55 - Sora Video 50
08:05 - Sora Video 51
08:15 - Sora Video 52
08:25 - Sora Video 53
08:35 - Sora Video 54
08:45 - Sora Video 55
08:55 - Sora Video 56
09:05 - Sora Video 57
09:15 - Sora Video 58
09:25 - Sora Video 59
09:30 - Sora Video 60
09:40 - Sora Video 61
09:50 - Sora Video 62
10:00 - Sora Video 63
10:10 - Sora Video 64
10:20 - Sora Video 65
10:30 - Sora Video 66
10:40 - Sora Video 67
10:50 - Sora Video 68
11:00 - Sora Video 69
11:10 - Sora Video 70
11:20 - Sora Video 71
11:30 - Sora Video 72
11:40 - Sora Video 73
11:50 - Sora Video 74
12:00 - Sora Video 75
12:10 - Sora Video 76
12:20 - Sora Video 77
12:30 - Sora Video 78
12:40 - Sora Video 79
12:50 - Sora Video 80
12:57 - Sora Video 81
13:07 - Sora Video 82
13:17 - Sora Video 83
13:27 - Sora Video 84
13:37 - Sora Video 85
Let's dive into the Kaiberr's AI Powerhouse, SuperStudio!
Whether you’re new to AI tools or a seasoned pro, this video explores the incredible creative potential of Kaiber’s canvas-based design. From generating hyper-realistic images to punk-inspired character concepts, I’ll walk you through the ins and outs of this groundbreaking AI platform.
⚡ Key Highlights
• Understanding the new canvas
• Exploring creative templates and Image Lab modules
• Generating stunning visuals and animations with audio-reactive features
• Concept building: from cyberpunk to steampunk, and even punk-punk!
🧠 Why Kaiber 2.0 is a Game Changer
Whether you’re creating magical punk rock bands or futuristic cityscapes, Kaiber’s ever-evolving platform delivers endless possibilities. Stick around as I showcase how the stencil feature can transform your characters and discuss the impact of its reintroduced audio reactivity.
LINKS:
KAIBER: https://kaiber.ai/superstudio?via=tim
KAIBER TEMPLATES - https://play.superstudio.app/creative-templates?via=tim
Previous Video On Kaiber: https://youtu.be/ZjsF64EFrHs
Chapters
0:00 Intro
0:59 Kabier's New Canvas:
1:37 Generating Stunning Images with Flows
02:29 - Generating Video In Kaiber
03:16 - Getting Weird with Kaiber and Arcane
03:45 - Creating a Character in Kaiber
04:10 - Blending Aesthetics in Kaiber
04:48 - Blending Moods in Kaiber
05:36 - Upscaling Images In Kaiber
06:11 - Using Kaiber as a Creative World Builder
07:07 - Filling out the rest of the Characters
07:27 - Posing Your Characters in Kaiber - Super Powerful!
08:42 - Animating Your Posed Characters
08:56 - Audio Reactivity In Kaiber
09:47 - Generating a Title Sequence In Kaiber
10:14 - Final Thoughts & Getting Started with Kaiber
RunwayML Just Dropped Two Game-Changing Updates!
RunwayML is shaking up the AI image and video scene with Video Expand and their very own image model, Frames! In this video, I dive into these exciting new features to see just how far they can push creative boundaries. Whether you’re looking to resize videos seamlessly with AI inpainting or create stunning images with unparalleled control, RunwayML is here to impress.
✨ Here’s What You’ll Learn:
What Video Expand is and how it turns 9:16 videos into 16:9 (or vice versa!)
Hands-on tests with Gen-3’s incredible inpainting abilities.
Exploring Runway’s new Frames image generator, built entirely in-house.
Creative experiments, unexpected results, and all the quirks that make this tech fascinating!
🕒 Chapter Timestamps:
00:00 Intro: RunwayML’s Big Updates
00:24 Feature 1: Video Expand Overview
01:12 Real-Time Tests: Stock Footage
01:48 - Expand Stock Footage Test
02:38 - Combining Video Expand with Text Prompts
03:44 - Changes to Footage
04:33 - Testing Expand with Old TV Shows
06:01 - The Shining Gen-3 Expand Test
06:49 - Community Outputs
07:29 Pushing Gen-3 Expand to Ultra Wide Screen
08:25 - The Dark Knight & Kill Bill Expanded to Ultra Wide Screen
09:15 - Using First Frame To Minimize Errors
10:10 Feature 2: Frames – Runway’s New Image Model
10:43 - Examples of Runway's New Image Model
AI tools just keep leveling up! 🚀 In today’s video, I dive into this week’s biggest updates in the creative AI space. Whether you’re into audio generation, image-to-video tools, or cutting-edge features in video models, there’s something here for everyone. Don’t miss out—use the chapter breaks below to jump to your favorite topics! 👇
0:00 Intro
0:18 Suno V4 - Audio quality updates, lyric generation, and new community tracks.
2:25 Flux Logo Tool - Slap a Logo on A Flux Image!
03:17 - Flux Tools: Inpainting and Editing Flux Images!
03:52 - Runway Gen-3 - Longer outputs, vert format, and advanced camera controls.
04:17 - LTX Studio- Exciting news on a NEW Video Model!
05:15 - New Minimax Controls!
05:39 - Freepik AI Video Is Now Live!
06:08 🖌️ Samurai: AI Rotoscoping - Motion-aware visual tracking.
07:11🏔️ Skyglass Virtual Sets - Cinematic shots with just your phone.
08:00 ⚡ Effect Texture Maker - Create dynamic VFX and animations.
09:24 Vidu 1.5 - Reference image combos for stylistic animations.
10:42 - Hotshot v2v with Avatars!
11:19 🤖 OpenAI News - Live camera updates
11:47 - Latest Sora Video and Release?
Links Mentioned in the Video
Suno: https://suno.com/
Glif: https://glif.app/glifs/cm3o7dfsd002610z48sz89yih
LTX Studio: https://ltx.studio
Flux Tools on Replicate: https://replicate.com/exp
Samurai Rotoscoping Tool: https://github.com/yangchris11/samurai
Skyglass App: https://www.skyglass.com/
Effect Texture Maker: https://mebiusbox.github.io/contents/EffectTextureMaker/
VIDU: https://www.vidu.studio/
HOTSHOT https://hotshot.co/
Can AI Fix the Polar Express? Today I'll dive in to see if we can use AI Tools such as Midjourney, and the various AI Video Generators to breathe new life into some legacy CGI films. Plus, we’ll talk about what’s possible right now, what challenges remain, and when this could become mainstream for platforms and studios.
LINKS
Making Your Own AI Film: https://youtu.be/hjk4a9gxILc
AI FaceSwapping: https://youtu.be/3pp7qw19nuA
Kai Turner's New Channel: https://www.youtube.com/@UCd5LOHXpyRAK2U6gUvHw3Mg
Chapters:
00:00 – Intro: Can We Fix The Polar Express?
00:23 – Why The Polar Express Needs AI Magic
01:34 - If you're bothered by this...
02:09 - Using AI to Facelift The Polar Express
02:39 - Midjourney's Retexture Feature
03:13 - Polar Express Retextured
03:38 - Video Generation via Minimax
04:19 - My Experiment with The Polar Express
04:49 - The "Trick" with AI Polar Express
05:32 - Preserving Performance in AI CGI Restoration
06:01 – Limitations and Challenges of AI Restoration
07:07 – Using Face Swaps to Add Realism
07:41 – When Will This Tech Be Primetime Ready?
08:02 - Why use AI Reimagining for Movies?
08:36 – Should Studios Use AI to Revamp Old Movies?
09:04 – What You Should Use this Tech For
Discover the latest in AI image and video generation with BaseLabs and Flora! 🚀 In this video, I'll dive deep into two innovative platforms that are reshaping the AI creativity landscape. From BaseLabs’ comprehensive suite featuring Flux, Stable Diffusion 3.5, Recraft, and image-to-video capabilities, to Flora’s experimental canvas-based workflows for building immersive worlds, we explore their unique offerings and how they stand out in a crowded market.
Both platforms offer free credits, so you can test them out yourself! I'll demonstrate how to generate personalized images using trained models, create videos from images, upscale your creations, and even build interactive stories with AI. Whether you’re a content creator, AI enthusiast, or just curious about the latest tech trends, this video has you covered.
As a note, I did reach out to Based Labs to score you guys a discount. This video isn't sponsored at all, I'm just trying to do that whenever I can to save you all some coin!
Don’t forget to like, comment, and subscribe for more AI tech updates!
LINKS:
BASEDLABS https://www.basedlabs.ai/?via=tim
(Use Code THEO15 for 15% off the first 3months)
FLORA: https://www.florafauna.ai/
Chapters:
00:00 - Introduction
00:24 - The Rise of New AI Platforms
00:52 - Exploring BasedLabs
01:36 - What Models are available on Based Labs?
02:16 - Image Generation & Model Training in BasedLabs
03:04 - Important Note About Saving on BasedLabs
03:51 - Editing Images in BasedLabs
04:01 - Image-to-video generation on BasedLabs
04:31: Base Labs - Model Comparison
04:44 - Comparing different models on BasedLabs
05:38 - Using trained models with LoRAs on BasedLabs
06:19 - Upscaling images on BasedLabs
06:44 - Pricing plans for Base Labs
07:05 - Introduction to Flora AI
07:19 - Overview of Flora's canvas-based workflow
07:42 - Using the "World Builder" example in Flora AI
09:32 - Creating character variations in Flora AI
10:10 - Using different image-to-video models in Flora AI
10:54 - Exploring other use cases for Flora AI
11:32 - Demonstrating image variations in Flora AI
12:12 - Pricing plans for Flora AI
I’m diving into the easiest way to generate consistent characters for AI Images and Video! Whether you want to create yourself or any other character, this method in Flux takes only a few clicks, and I promise, it’s really that easy!
I’ll walk you through Krea's latest feature for training characters (or objects) with low-ranking adoption (LoRA) in Flux. I’ll also be comparing Krea’s approach with the new consistent character training feature in Kling, which I covered in my last video. Expect a fun head-to-head showdown between Minimax and Kling as I put them to the test!
Chapters:
0:00 – Intro & What You’ll Learn
0:27 – Why Krea is a Silent Killer in Creative AI
01:16 - Previous Training with Kling
01:52 – Training Flux in Krea
02:21 – Setting Up Your Own Character with Photos
03:37 – Adjusting Styles and Trigger Words for Precision
04:13 - First Results
05:01 - Adding Styles for Better Results
06:19 - Really bringing it together
06:58 - Experiment to hone in
07:40 - Once you Have a Style Locked In
08:14 – Minimax vs Kling Shootout
09:19 - Flipping the Test with Kling
10:18 – Experimenting is the Key
11:31 – AI vs. Human Screenwriter Results!
12:27 – Synthetic Characters in Kling 1.5
13:09 – NYC Alt Studio Meetup Info
Mentioned Links & Resources:
Krea: https://www.krea.ai/home
Previous video on Kling’s character training: https://youtu.be/UnSJ7W8A7c4
Meetup link for NYC Alt Studio Community Event: https://lu.ma/zmzfxfib
Come hang out if you’re around NYC! As always, remember that this is the worst AI will ever be—only up from here! Thanks for watching!
Dive into the future of AI video creation with Kling's groundbreaking new feature! Witness mind-blowing character face-swapping and training capabilities that let you become anyone, anywhere.
In this video, we explore:
The incredible realism and ease of use of Kling's AI Video model.
A step-by-step guide to training your own AI character.
Insider tips and tricks for optimal results.
A glimpse into the future of AI video technology.
LINKS:
KLING: https://klingai.com
TechHalla: https://x.com/techhalla
DJI Pocket 3 (I swear, I'm not sponsored by them! hah) https://amzn.to/3CeDHT1
CHAPTERS:
0:00 Introduction and Overview of Kling's New Feature
0:28 Kling Early Beta
0:54 Examples of AI Character Transformations
1:32 Comparison with Deepfake Technology
2:39 Training Your Own AI Model in Kling - Step-by-Step
3:42 Tips for Shooting Optimal Training Videos
4:32 Additional Tips
5:29 After you have Trained
5:48 Generating AI Videos with Your Custom Character
6:42 Exploring TechHalla's AI Character Examples
7:58 Character Limitations and Bleed
8:22 Tips on Model Bleed
8:51 - Text to Video is still Weird
9:18 The Evolution of AI Video and What's to Come
Today we’re diving into the rise of Red Panda, the mysterious image generator that has overtaken giants like Flux 1.1 Pro and Midjourney 6.1. Is it really a game-changer, or just a cute codename?
Also, we're taking a look at a new method of AI Image compisition from Blockade Labs. This is something I've been waiting for, and its a TON of fun!
Also, a big thanks to today's Sponsor: Hostinger! Sign up to build a website and get 10% off with my coupon code: THEORETICALLY
http://hostinger.com/theoretically
🔗 Related Links & Resources:
Recraft Model Platform: https://www.recraft.ai/invite/eOAzSolq8z
Artificial Analysis Leaderboard: https://artificialanalysis.ai/
Blend Box & Skybox by Blockade Labs: https://www.blendbox.ai/
Chapters:
00:00 - Intro: The Rise of Red Panda
00:59 Where do ELO Scores Come From?
01:55 - Red Panda Revealed!
02:22 - Other quick canvas updates
02:43 Recraft AI First Impressions
04:28 Trying out some more imaginative stuff: cyberpunk woman with white hair
05:37 Recraft AI Color Adjustment
05:47 Recraft AI Text Generation
06:45 Community Outputs
07:25 Recraft AI Vector Art
07:47 More Community Outputs
08:23 Recraft AI Pricing
08:45 - Build a Website with AI
09:23 Hostinger Website Builder Walkthrough
12:55 Blockade Labs Blend Box
13:23 Blendbox Walkthrough
14:44 Where this is going to get super powerful
Don’t forget to like, subscribe, and hit the bell icon to stay updated with more AI tech news and tutorials!
The creative AI world is exploding with new models, updates, and even a showdown between human and AI screenwriters! Let's dive deep into the latest advancements, including:
Mystic 2.5: A powerful new image generator integrated into Magnific, offering stunning 4K resolution and impressive creative control.
PixVerse V3: This text-to-video model boasts interesting camera movements and diverse stylization options, but how does it stack up against the competition?
A Contender to Minimax? A surprising contender emerges with impressive video generation capabilities. Could this be the new Minimax?
Human vs. AI Screenwriter: Can you tell the difference between a script written by a human and one generated by AI? Put your skills to the test with this intriguing story from No Film School!
Chapters:
00:00 Introduction
00:24 Mystic 2.5
01:14 Mystic 2.5 Tests
02:50 - A new Character Appears!
03:24 - Creativity in Mystic 2.5
04:04 Model Comparison
06:43 Is It Flux Under the Hood?
08:02 Pixverse V3
08:42 New Pixverse Features
09:33 Image to Video
10:04 Text to Video
10:26 - 3d Model in Pixverse
11:25 Really stellar Pixverse output
12:39 Famous Cinematic Shots
13:10 Community Outputs
14:01 A New Minimax Contender?
15:31 Testing it out
16:20 Human vs. AI Screenwriters
Links:
Mystic by Magnfic: https://magnific.ai/
PixVerse: https://pixverse.ai/
Seller Pick: https://www.sellerpic.ai/
No Film School Website: https://nofilmschool.com/ai-screenwriting-challenge
Script A (Dropbox): https://roar-assets-auto.rbl.ms/files/68427/Script%20A.pdf
Script B (Dropbox): https://roar-assets-auto.rbl.ms/files/68473/Script%20B.pdf
Don’t forget to like, subscribe, and hit the notification bell for more exciting AI updates and experiments!
Conectar
Theoretically Media
