Uploaded December 2025 | Updated November 2026, 4 hours ago
We are closing out the year with a massive update from Luma Labs! Today, I’m taking a deep dive into the new Ray 3 Modify feature. This isn't just a simple filter—we're looking at keyframing, start/middle frames, and character consistency that might just change your workflow.
I’ll show you how to fix "broken" AI generations, remove unwanted objects (like boom mics), and restyle footage completely. I'll also share a "stupid" but essential trick for removing characters that requires reversing your footage.
Plus, we have a mystery on our hands. A new model codenamed "Sicily" has appeared in the arena—is this our first look at Google Veo 3.5 or even Veo 4? We also check in on ByteDance's wide release of SeaDance 1.5 Pro and the new realism update from Topaz Labs.
Let's break it all down before we head into 2026!
Links Mentioned:
Luma Labs Dream Machine: lumalabs.ai/dream-machine
ByteDance SeaDance: fal.ai/models/fal-ai/bytedance/seedance/v1.5/pro/text-to-video
Topaz Labs Video: topazlabs.com/ref/2518
00:00 Intro: No Slowing Down
00:34 Luma Ray 3 Modify Overview
01:10 The Modify Interface Explained
01:40 Understanding the Strength Slider
02:14 Start Frames vs Modify Frames 02:31
Easy Mode: Fixing Dodger Stadium
02:53 Easy Mode: The Green Tuxedo Fix
03:24 Removing Boom Mics from Footage
04:02 The "Reverse Footage" Removal Trick
04:47 Stylization Test: Dark Angel
05:42 Stress Test: The Fight Scene
06:26 Attempting Photorealism (Mortal Kombat)
07:04 Slow Motion Handshake Test
07:28 The "Middle Frame" Workflow
08:01 How Luma Batches Generations
08:41 Character Reference Returns (Ripley)
09:26 Pushing the Slider for Creativity
09:48 The Lip Sync Problem
10:19 Narrative Test
11:07 Verdict on Luma Modify
11:24 Mystery Model: Is Sicily Google Veo?
12:17 Comparing Sicily to Veo 3.1
12:40 Google Veo 4 Predictions
12:58 ByteDance SeaDance 1.5 Pro Release
13:39 SeaDance Native Audio Test
14:15 Topaz Labs: Bloom & Astra Update
15:01 Testing Topaz Starlight Precise
15:26 The "Garbage In, Garbage Out" Rule
16:14 Holiday Wishes & 2026 Outlook
We are closing out the year with a massive update from Luma Labs! Today, I’m taking a deep dive into the new Ray 3 Modify feature. This isn't just a simple filter—we're looking at keyframing, start/middle frames, and character consistency that might just change your workflow.
I’ll show you how to fix "broken" AI generations, remove unwanted objects (like boom mics), and restyle footage completely. I'll also share a "stupid" but essential trick for removing characters that requires reversing your footage.
Plus, we have a mystery on our hands. A new model codenamed "Sicily" has appeared in the arena—is this our first look at Google Veo 3.5 or even Veo 4? We also check in on ByteDance's wide release of SeaDance 1.5 Pro and the new realism update from Topaz Labs.
Let's break it all down before we head into 2026!
Links Mentioned:
Luma Labs Dream Machine: lumalabs.ai/dream-machine
ByteDance SeaDance: fal.ai/models/fal-ai/bytedance/seedance/v1.5/pro/text-to-video
Topaz Labs Video: topazlabs.com/ref/2518
00:00 Intro: No Slowing Down
00:34 Luma Ray 3 Modify Overview
01:10 The Modify Interface Explained
01:40 Understanding the Strength Slider
02:14 Start Frames vs Modify Frames 02:31
Easy Mode: Fixing Dodger Stadium
02:53 Easy Mode: The Green Tuxedo Fix
03:24 Removing Boom Mics from Footage
04:02 The "Reverse Footage" Removal Trick
04:47 Stylization Test: Dark Angel
05:42 Stress Test: The Fight Scene
06:26 Attempting Photorealism (Mortal Kombat)
07:04 Slow Motion Handshake Test
07:28 The "Middle Frame" Workflow
08:01 How Luma Batches Generations
08:41 Character Reference Returns (Ripley)
09:26 Pushing the Slider for Creativity
09:48 The Lip Sync Problem
10:19 Narrative Test
11:07 Verdict on Luma Modify
11:24 Mystery Model: Is Sicily Google Veo?
12:17 Comparing Sicily to Veo 3.1
12:40 Google Veo 4 Predictions
12:58 ByteDance SeaDance 1.5 Pro Release
13:39 SeaDance Native Audio Test
14:15 Topaz Labs: Bloom & Astra Update
15:01 Testing Topaz Starlight Precise
15:26 The "Garbage In, Garbage Out" Rule
16:14 Holiday Wishes & 2026 Outlook
![Octo Is Seedance’s Prompt Killer
Dreamina just rolled out Octo, a new canvas/agent workflow that tries to turn Seedance 2.0 into something closer to a scene builder than a prompt box.
So I tested it with character sheets, storyboards, a cyberpunk fight sequence, and a timeline extension to see whether Octo actually makes AI video workflows easier, or just moves the chaos into a new interface.
We’re also looking at NVIDIA Cosmos-3, Black Forest Labs showing Flux to Martin Scorsese, Bernini as an open-source “Google Omni for video” style model, and TripoSplat for turning 2D images into controllable 3D-ish spaces.
Basically: AI video is getting less like “write one prompt, get one clip” and more like planning, editing, world-building, and trying to keep the machine pointed in the right direction.
Big thanks to #genspark for Sponsoring Todays Video. You can get Free Credits at Genspark by using this link: https://www.genspark.ai/?utm_source=yt&utm_campaign=TheoreticallyMedia02
#WorkWithGenspark
Links:
Dreamina / Octo: https://dreamina.capcut.com/ai-tool/home?utm_source=Officiaaccount&utm_campaign=octo&utm_content=6ytb
NVIDIA Cosmos-3: [add link]
Cosmos-3 on FAL:
Black Forest Labs / Scorsese clip: https://x.com/bfl_ai/status/2061800769226252721?s=20
Bernini: https://bernini-ai.github.io/
TripioSplat: https://huggingface.co/spaces/VAST-AI/TripoSplat
CHAPTERS:
00:00 AI video workflows are heating up
00:24 Dreamina launches Octo
01:01 Using Octo as a canvas agent
01:51 Character sheets and storyboards
02:28 Where Octo starts to struggle
03:14 The Seedream / image model problem
04:03 Testing Octo with a cyberpunk character
04:28 The agent brain feels underpowered
05:27 Fixing the character sheet manually
05:56 Building the scene anyway
06:58 Turning the elements into video
07:35 Octo’s timeline extension
08:27 The part Octo actually nails
08:38 Is this the future of AI video workflows?
10:16 NVIDIA Cosmos 3
11:38 NVIDIA wants to be the physics department
12:21 Martin Scorsese meets Flux
13:03 Sponsor - GENSPARK
15:21 Bernini, an open-source Omni-style video model
16:23 How Bernini plans video edits
17:17 The H100 problem
17:49 Triple Splat turns images into 3D spaces
18:32 The cheap World Labs workaround
19:07 Wrap-up
#seedance2, #nvidia, #opensourceai Octo Is Seedance’s Prompt Killer](https://i.ytimg.com/vi/EmzS3MiXLfU/mqdefault.jpg)









