Ai and things
Duckstown
Joined October 2021
- Tweets5K
- Following3.9K
- Followers3.7K
- Likes8.3K
Partymoji skill creates a music video with emojis that reacts to your music.
Blender 5.x with the Blender MCP add-on connected to Claude
A Descript connection (used for lyric timings)
github.com/MayorDapper/Party…
Ducko retweeted
Blender just became 10x more powerful with AI. 🚨
Blender + GPT Astra = a completely different way of creating visuals.
Instead of choosing between traditional 3D and generative AI, you can combine both.
→ Build your scene in 3D
→ Control the camera, depth, and composition
→ Use AI to transform the result
→ Create visuals that would normally take hours or days
This is where creative workflows are heading.
Comment “LAB” and I’ll show you how to do this.
Ducko retweeted
Restoring old footage with AI, the full breakdown
I tested two things for this and both gave good results, @higgsfield_ai Genjutsu and @Hailuo_AI MiniMax H3.
Steps and prompts below 👇
🤖 Made with AI
Anyone tried this yet? It works with Astra but wondering if it will work fully local.
You ────────────────┼── DeepSeek Harness
│
└── Qwen Code
│
▼
existing ltx-2-mlx
│
▼
LTX-2.5 MLX Q8
│
▼
M series Macs
1
86
Candid observation: All from the same prompt.
Competitors:
Seedance 2.5
Googe Omni
Grok Imagine
Lets go!
First is Seedance 2.5
1
1
119
Ducko retweeted
Seeing what people are making with ASTRA and Blender MCP got me wondering how much of my own filmmaking workflow I could handle through conversation.
So I asked ASTRA to build me a 3D previs tool, and we talked through everything until generation and editing.
Yes, “we”—me and my agents.
We started with an original idea and developed the story, script, beats, and storyboard together.
I described the shots, watched them play out, and kept talking: faster chase, closer camera, longer transformation. New ideas came up while we were fixing things.
We built the character sheets, prompts, and the harness around the process: camera rules, continuity checks, and generation checks. The project memory and working context stay local.
I only hit generate three times for the video footage.
It’s not perfect. The point was to see whether this workflow could take us from an idea to footage I could edit into a film.
This comparison shows the previs, alongside the geography, distance between the characters, and character sheets.
I don’t want my workflow tied to one subscription or spend my day dragging nodes around. I’d rather talk, make things, and brainstorm as we go.
Eventually, I want an agent that knows what I mean by “more tense,” remembers what we’ve tried, and understands which details matter to me.
The more we make together, the less I should have to explain from scratch.
And I think GPT6 ASTRA is just the beginning, we need to be prepared for what’s coming.
made with @atlas_cloud_ai API with Seedance 2.5
#AtlasCloud
Getting my hands dirty with Astra
Pretty neat for building apps and sites. And NO we have not hit AGI yet. But the model is a hit, no doubt.
1
1
80
Ducko retweeted
Fable 5 does crazy things on Supercomputer.
A Higgsfield creator built an app that blocks 3D scenes, records a real cinema camera move, and renders the video through Seedance 2.0.
A 15-second clay pass carries the exact trajectory, timing, and framing a prompt keeps losing.
Tested the Grok Imagine 1.5 Preview.
Physics is fantastic but to be honest the voice dialogue between two characters is still a mess. Two people in conversation saying the same words at the same time still happens, the man says the womans lines regardless of prompt direction....
It would have been a fail if its not in Beta. Still testing.
1
89
Ducko retweeted
Generate split-stack videos like this in a single click.
The basic setup is simple, just upload an image and the system takes care of the rest.
Want more control? Switch to graph mode and fine-tune each step to dial in exactly the look you’re after.
Click the link below to use the workflow 👇
WAN 2.7 has a glitch.
If your song says anything like, "You make me feel high", the character starts behaving intoxicated when you try lipsync and the screen starts to go fuzzy and smokey. Its flipping insane.
2
53
Ducko retweeted
AI Art Is Already the Future
And the Future Is Winning
Let’s stop pretending.
AI art isn’t “coming.”
It’s here. It’s loud. It’s everywhere.
And it’s only getting better.
While some people are still screaming “not real art,” thousands of artists are quietly building careers, getting brand deals, winning clients, and creating work that moves millions.
The numbers don’t lie. The platforms don’t lie.
The audience doesn’t lie.
We are not stealing art.
We are not replacing artists.
We are the new generation of artists, faster, freer, and unafraid of tools that previous generations could only dream of.
The gatekeepers can keep clutching their brushes and calling it “soul.”
Meanwhile, we’re out here making soul at scale.
The present belongs to those who create.
The future belongs to those who adapt.
To every AI artist still getting hate in the comments: Keep shipping.
Keep experimenting.
Keep proving them wrong with pixels.
The renaissance isn’t coming.
It’s already here, and we’re the ones painting it.
Who’s with me?
Ducko retweeted
Fun interactive science app ideas | Part 3
Played around with generating 3D biological structures and made an app to explore them interactively
UI Design
GPT Images 2
Code
Gemini 3.1 Pro
More demos ↓
