Sherm Studios retweeted
Save Jensen!
Jensen Huang's exit from Nvidia was marked by corporate scandal, but many of his own former employees and colleagues across Silicon Valley are coming out in droves to express support for the man they believe to be a legend. cnn.com/2026/11/15/business/…
Sherm Studios retweeted
Six ways to use Seedance 2.5, all on one very grumpy stop-motion cat.
These are the tips I wish I'd known when I started. A quick tour of what each mode can do and how to drive it. 🧵
🚨BREAKING 🚨
Jeff Bezos’ transformation, from a bald-headed man to someone with a full head of hair, will always be a surprise. Lauren Sánchez must be so happy with the transformation.
🤖 Made with AI
297
I've started making film stills on my local Mac Studio. No expensive credits. Just local renders!
To test the capabilities I ran the same prompt and seed through 3 local image models.
Z-Image-Turbo looked the most real to me. It's on top, FLUX.3 in the middle, Qwen-Image 2.1 on the bottom.
See my process and steal my prompt below 👇
1
1
79
My setup:
- Mac Studio Ultra, everything local
- mflux by Filip Strand, which runs these models on Apple Silicon
- Z-Image-Turbo by Alibaba's Tongyi lab, 9 steps, 1536x640, about 26 GB of memory
1
28
Want to use this prompt? Copy it below 👇
35mm anamorphic motion picture film frame, photoreal live action, fine film grain. One infantry courier in a blue wool coatee reads a folded blank dispatch inside a farmhouse. A single high window at frame left supplies the only light, sending a hard shaft through flour dust with no fill. His close face is divided between pale window light and deep black shadow; the window recess makes a frame within the frame. Natural colors have a cool winter cast, without flare or split-color grading. Exactly one person is in frame, holding one unmarked paper; no readable text appears.
6
Sherm Studios retweeted
We're creating a remastered version of Battlefield: Vietnam since nobody else is gonna do it.
Retweet if you want a key to the closed beta.
The GOAT
My goal with AI is to make it ✨i n v i s i b l e✨
Every decision I make, from prompting to selecting which generation actually works, is in pursuit of this aim.
AI will always have imperfections, but the one thing I refuse to compromise on is whether my characters are giving a believable performance or not.
Because imo, that's the one thing that can make people forget they’re watching AI at all.
Take the attached clip for example:
Watch Karl right after Beethoven says “Well?” He tosses his pen back into the holder and gives a look that says “brother, can you PULEAASE stop talking!”
That's something that I prompted for specifically.
[He tosses his quill down, irritated, and turns his head to the side to look at his brother, a little puzzled at what has got into him, and answers him. He asks line 1 as a real question, dry, mildly exasperated, as if to say, "What's up with you?"]
Those details were enough to yield a performance that, to me, really felt like an older brother pestering his little brother.
And notice how I'm saying "to me."
Because I can't always tell you why I'm choosing one take over another. The reasoning behind my choice could very well come down to something as small as “he didn’t exhale fast enough.”
I'm just letting my intuition be my guide.
I could hand you my exact prompts, character refs and environments, and you’d still end up with a different scene from mine due to a difference in our intuition.
But that's kinda the point, isn't it?
Anyone can copy prompts and JuSt pUSh A bUtTOn until thy kingdom come, but knowing which takes to keep is the actual craft behind all of this.
1
4
94
I made "ISLA", a 58 second AI short film about my cousin Chris who passed away a few years ago.
I made it in one evening, and learned a lot along the way.
One of the shots where I wanted the heather falling into her dad's grave past the camera, took 4 different takes to regenerate.
Here are all 4, with exactly what I changed in each prompt.
8
6
323
What I learned:
1. If the image model won't make the still with a person in it, make the empty set and let the video walk them in.
2. Give timings in seconds.
3. Use a reference photo for characters instead of visually describing it. You'll have better character consistency as you generate more and more shots.
4. Sometimes the fix isn't a better prompt. It's a different camera idea.
1
14
Want to use this prompt? Copy it below 👇
One single continuous shot, no cuts, in slow motion, dreamlike and weightless, a quiet and respectful moment of remembrance. It opens on the frame: the girl's small hands holding the sprig of purple heather above the open grave, her black wool coat behind them. Her fingers open and the heather slips free. The camera follows the heather down as it falls in slow motion, floating and turning gently, drifting a little from side to side like a feather on the air, down past the dark earth walls into deepening darkness, the soft grey light fading above. Very shallow focus: only the heather is sharp, everything else melts into soft darkness. Near the end the camera slows and the heather drifts right past the lens in a soft purple blur, leaving darkness. No words are spoken. A soft rush of air, the faint rustle of the heather, a low hollow hum, no music, no score. 35mm film, vintage anamorphic lens, natural light, real skin texture.
Cinema Studio 4.0, 10 seconds, start frame: her hands holding the heather.
38












