1:25
Erinys (YC S26) - Launch Video
Erinys · B2B · Aug 2026
Explore
Dyna Robotics - Today we are introducing Dyna-2, a world-action model pre-trained on
Dyna · Aug 2026 · 1:42 · AI-annotated
- Length
- 1:42
- Scenes
- 29 · 3.5 s avg
- Product on screen
- 0:16
- Features shown
- 5
- Voiceover
- No
00:00:00 / 01:42:08
Dyna’s 1:42 launch film runs 29 scenes, about 3.5 s each. The product shows up at 0:16 and 5 features are revealed. There is no explicit call to action. The video alternates flat beige statement cards, near-black self-drawing chart cards, and dotted-frame live-action grids, moving from a task montage to a scaling-law proof and then to labeled feature demos. Signature moves are the zoom-out mosaic that assembles the Dyna-2 wordmark from a wall of generated clips, the green line that draws itself across each chart, and the split-screen mirroring of a human and the robot.
Footage neededEstimated needs: 1 official logo — reused across the film · 17 filmed passages — shoot, supply or replace · 5 graphic passages — draw or supply
Is this your launch? to your site or press kit.
More like this
1:09
Cal.com introducing Dynamic Group Links
Cal.com · Productivity · Aug 2022
1:07
Deploy (YC F26) - The Generalist Robot Deployment Company
Deploy · Industrials · Oct 2026
Dyna Robotics - Today we are introducing Dyna-2, a world-action model pre-trained on, scene by scene
· 29 scenes as text
01 · Hook · 0:00–0:02
A dim frame brightens to show a two-armed robot at a wooden board, one gripper holding a knife, the other a celery stalk, a tray of chopped pieces below. White text One Model. holds at lower left.
On-screen text: “One Model.”
Camera: Push in
02 · Hook · 0:02–0:03
From black, a dim shot reveals a robot arm holding a knife over a celery stalk on a wooden board, chopped pieces in a metal tray, then the image fades back to black. The caption stays locked in the lower left throughout.
On-screen text: “One Model.”
Camera: Static
03 · Hook · 0:03–0:07
A split screen pairs two robot demos: on the left, a black-gloved hand lifts a neon highlighter from a clear desk organizer, while on the right a metallic arm reaches into a glass fridge stocked with drinks. The caption One Model. holds at the lower left.
On-screen text: “One Model.”
Techniques: Split screen
Camera: Static
04 · Hook · 0:07–0:09
A 2x2 grid of live-action clips shows a two-gripper robot arm on a white table: it coils a green cable, drops a screwdriver into a cardboard box, works on a white car door, and lifts a teal laptop lid. The caption “Hundreds of tasks” settles into the bottom left panel.
On-screen text: “Hundreds of tasks”
Techniques: Split screen
Camera: Static
05 · Hook · 0:09–0:10
A 2x2 grid of four live-action clips shows robotic grippers performing different tasks, handling a green cable, a tray with tools, a car door panel, and a teal sheet, while a two-line caption sits over the bottom-left panel.
On-screen text: “Hundreds of tasks”
Techniques: Split screen
Camera: Static
06 · Hook · 0:10–0:13
A 3x3 grid of clips plays at once, each showing a white robot arm with a black gripper doing a task: placing a can, packing a suitcase, folding fabric, opening a drawer, opening a box, sorting snacks. The caption “Hundreds of tasks” holds over the bottom-left clip.
On-screen text: “Hundreds of tasks”
Techniques: Split screen
Camera: Static
07 · Problem · 0:13–0:16
On a flat beige background with four small corner dots, the line 'Hundreds of tasks' crossfades out as 'How do we scale it?' fades in at the center, holds, then fades out.
On-screen text: “Hundreds”, “of tasks”, “How do we scale it?”
Techniques: Kinetic type
Camera: Static
08 · Product intro · 0:16–0:24
A grid of first-person clips of hands cooking, packing and cleaning multiplies into a dense mosaic of hundreds of tiny clips, a white caption appears at the bottom, then the view zooms in as the tiles brighten into the wordmark Dyna-2.
On-screen text: “All individual clips generated by Dyna-2”, “We trained a model with 1 million+ hours of human data”, “Dyna-2”
Techniques: Masking
Camera: Pull out
09 · Product intro · 0:24–0:27
A dense mosaic of tiny video clips fills the frame, with the word Dyna-2 spelled out in large white tiles at the center. White captions appear at the bottom, first about training data, then about scaling laws, while a small credit sits at the lower left.
On-screen text: “All individual clips generated by Dyna-2”, “We trained a model with 1 million+ hours of human data”, “and discovered surprising scaling laws...”
Camera: Static
10 · Product intro · 0:27–0:30
A dark wall of tiny moving video clips fills the frame. The word Dyna-2 is spelled in bright pixelated letters across the center, each pixel a clip. A white caption reads below, with a small monospace credit at the bottom left. The frame is locked off.
On-screen text: “and discovered surprising scaling laws...”, “All individual clips generated by Dyna-2”
Camera: Static
11 · Product intro · 0:30–0:31
A dark wall of tiny generated thumbnails fills the frame. The brand name Dyna-2 sits centered in large white mosaic letters and scales up. A gray caption below fades out, and a small monospace credit stays pinned to the bottom left.
On-screen text: “Dyna-2”, “and discovered surprising scaling laws.”, “All individual clips generated by Dyna-2”
Techniques: Kinetic type
Camera: Static
12 · Product intro · 0:31–0:32
A gray field holds the word Dyna-2, its block letters built from a dense mosaic of tiny video stills, with a small caption at the lower left. The word scales and dissolves as the background washes to off-white.
On-screen text: “Dyna-2”, “All individual clips generated by Dyna-2”
Techniques: Kinetic type
Camera: Static
13 · Feature reveal · 0:32–0:42
A dark chart card assembles: grid, axes and a legend box fade in, then a green line with four labeled points draws across, followed by a dotted fit curve, under a white headline. The card fades to black.
On-screen text: “For the first time, we discovered a scaling law for 1-million hours of human data”
Camera: Static
14 · Feature reveal · 0:42–0:51
A headline types in above a dark grid chart on a black card. A green line with error bars draws left to right across the plot, a dotted fit curve follows, then the whole card fades to black.
On-screen text: “This scaling law on human data implied a scaling law on never-seen robot data”
Techniques: Kinetic type
Camera: Static
15 · Feature reveal · 0:51–0:57
A dark chart card builds on a light background. A gray-to-green line draws from top-left to bottom-right as four data points pop in, a badge fades in, and dotted lines frame a -42% ERROR callout while the headline types out above.
On-screen text: “Video data is the new scaling axis; world modeling unlocks cross-embodiment transfer and can alone improve scaling”, “+1,000 VIDEO HR → -42% ERROR”
Techniques: Kinetic type
Camera: Static
16 · Feature reveal · 0:57–1:01
A split screen shows a woman in a white sweatshirt placing a blue folder into a binder on the left, while a white two-armed Dyna robot imitates the motion on the right. A gray pill label with Teachability fades in at the top right as the robot closes the folder.
On-screen text: “Teachability”
Techniques: Split screen
Camera: Static
17 · Feature reveal · 1:01–1:03
A split screen shows a woman in a white sweater with wrist sensors on the left, her hands moving over a binder, while a two-armed white robot on the right mirrors the gesture. A grey Teachability tag sits top right over a black-and-white patterned wall.
On-screen text: “Teachability”
Techniques: Split screen
Camera: Static
18 · Feature reveal · 1:03–1:05
A split screen pairs a woman with wrist sensors flipping a binder page on the left with a white two-armed robot copying the same flip on the right. She watches from behind the robot. A gray label reading Teachability sits top right.
On-screen text: “Teachability”
Techniques: Split screen
Camera: Static
19 · Feature reveal · 1:05–1:07
A 3x2 grid of live-action clips shows a robot arm with green grippers working at a fridge, each panel playing a different angle at once. A frosted caption box sits in the center, its quoted command swapping as the demos change.
On-screen text: “Instruction following:”, ““Open the fridge and grab the chocolate protein shake””, ““Open the fridge and grab the still water””, ““Open the fridge and grab the sparkling water””
Camera: Static
20 · Feature reveal · 1:07–1:08
A grid of six live-action clips shows a robotic arm opening a fridge and reaching for bottles, each panel playing a different angle at once. A translucent gray banner in the center holds the instruction text, static over the moving footage.
On-screen text: “Instruction following:”, ““Open the fridge and grab the sparkling water””
Techniques: Split screen
Camera: Static
21 · Feature reveal · 1:08–1:13
Eight live-action clips in a 4x2 grid show a robot arm opening a glass fridge and grabbing drinks. A translucent gray bar centered over the grid holds the caption “Instruction following:” with the quoted task swapping from Red Bull to sparkling water to Coke.
On-screen text: “Instruction following:”, ““Open the fridge and grab the Red Bull””, ““Open the fridge and grab the sparkling water””, ““Open the fridge and grab the Coke””
Techniques: Split screen
Camera: Static
22 · Feature reveal · 1:13–1:14
A two-by-three grid of six live-action clips shows a robotic arm opening a glass fridge and grabbing bottles. A translucent gray banner sits in the center with a heading and a quoted instruction that switches from vanilla protein shake to sparkling water.
On-screen text: “Instruction following:”, ““Open the fridge and grab the vanilla protein shake””, ““Open the fridge and grab the sparkling water””
Techniques: Split screen
Camera: Static
23 · Feature reveal · 1:14–1:19
A split screen frames a robotic arm with green grippers and a knife slicing a celery stalk on a wooden board. The right panel shifts through purple, pink, orange and green light while the left panel dims to black.
On-screen text: “Instruction following:”, “Open the fridge and grab the vanilla protein shake”, “Robustness and Precision”
Techniques: Split screen
Camera: Static
24 · Feature reveal · 1:19–1:20
Split screen: a robot arm slices a green stalk into even segments on a wooden board. The left panel fades to black while the right panel's lighting shifts from green to pink, with a caption held at the bottom.
On-screen text: “Robustness and Precision”, “SHARPIES”, “HIGHLIGHTERS”
Techniques: Split screen
Camera: Static
25 · Feature reveal · 1:20–1:26
A black-gloved robotic hand pulls the bottom drawer of a three-tier clear organizer, lifts a yellow highlighter from it and lays it on the wooden desk. A 'Dexterity' badge holds in the top-right corner through the locked-off low-angle shot.
On-screen text: “Dexterity”
Camera: Static
26 · Other · 1:26–1:33
A plain beige field holds a centered line that fades and scales in, then a second line fades in beneath it a few seconds later, both in thin dark gray type.
On-screen text: “One million hours isn't the limit.”, “It's the beginning of a new era of scaling.”
Techniques: Kinetic type
Camera: Static
27 · Other · 1:33–1:39
A light beige background holds a centered gray sans-serif question, the second clause set in a lighter shade. The line fades out as the dark gray DYNA wordmark and a small dot cluster scale in as the sign-off.
On-screen text: “Where does intelligence go, when the ceiling is gone?”
Camera: Static
28 · Outro · 1:39–1:41
A black cluster of dots and the uppercase wordmark DYNA sit centered on a flat pale beige field. The composition holds still while a slow scale settles the frame.
On-screen text: “DYNA”
Camera: Static
29 · Outro · 1:41–1:42
The DYNA logo, a black dot-cluster mark beside the wordmark, sits centered on a plain field that fades from light gray to black. The lockup scales slightly as the background darkens until the mark blends out.
On-screen text: “DYNA”
Techniques: Kinetic type
Camera: Static