each::labs

Close The Imagination Gap

[each::labs @eachlabs]

Introducing Lockstep.

Our new frontier model, and right now the best recast model on the market at the best price of any alternative.

Upload a video, add your characters, and every move stays locked. Fine-tune with an optional prompt, pick your resolution and scale to high-volume production.

Now on each::labs.

[MiniMax (official) @MiniMax_AI]

Start with a sketch. One product, every scene. Exactly as you prompt it.

Studio shot to lifestyle campaign, consistent in every frame. That's MiniMax H3 Image for e-commerce and ad creative at scale.

Thanks @eachlabs for showing it so beautifully ♥️ #H3Image

[each::labs @eachlabs]

Install it in Claude Code

/plugin marketplace add eachlabs/plugin
/plugin install eachlabs@eachlabs

Or in Codex CLI
codex plugin marketplace add eachlabs/plugin
codex plugin add eachlabs@eachlabs

Claude: https://t.co/d3NL9x8pqe
Codex: https://t.co/59MfMHb7eJ

[each::labs @eachlabs]

Your next Christmas ad can start with one prompt in Claude 🎄

Describe the ad you have in mind and the each::labs plugin creates it right in your chat. You don’t need to open another tool or pick a model yourself.

The each::labs plugin is waiting for you on Claude and Codex, with 600+ image, video and audio models.

[each::labs @eachlabs]

Scan the QR code to get early access 👀

[each::labs @eachlabs]

What a night, London 🍻

Thanks to everyone who came out to Bloomsbury with us and @vitpepper_stds. It was great to finally meet so many people building films, ads, games and apps with AI, and to hear about what you’re working on.

We also showed our new product for the very first time.

Want to be one of the first to try it? Check the next tweet ⤵

[each::labs @eachlabs]

Text to Image: https://t.co/JMt3L4byEc

Image to Image: https://t.co/45cYuTBEHn

Text to Image (High Speed): https://t.co/oAQaHVglmW

Image to Image (High Speed): https://t.co/PH9hQodfYf

[each::labs @eachlabs]

Native 4K images in about 10 seconds ⚡

MiniMax H3 Image, the new open-weight model by @MiniMax_AI, is now on each::labs.

It generates at up to 16MP in any aspect ratio, takes up to 16 reference images and edits with precision, from adding or swapping objects to rendering text, infographics and UI layouts.

Try it now on each::labs⤵

[each::labs @eachlabs]

Beat everyone to Vice City 🌴

Upload one selfie and the Vice City Collage workflow makes you its star, in one high-res collage you can post before the countdown even ends.

One selfie, one click. Live now on each::workflows.

🔗 https://t.co/frDi7kLCWH

[each::labs @eachlabs]

Your agents can run on Claude Fable 5.2 today.

It formalized the proof of Fermat’s Last Theorem in 30 hours with 1,000 agents. With the same setup, Opus 5.5 got 26% of the way in 18.

Change one model string and try it on each::router.

🔗 https://t.co/Ze7LczUMwf

[each::labs @eachlabs]

https://t.co/0IWvk29eDi

[each::labs @eachlabs]

You see every trend at its peak, but get stuck making it yourself? ☕

This one’s the Fancy Café First Time Video. Add your photo and you’re the one sitting proudly on the tiny table, catching the side-eye and quietly switching seats.

Every trend is one click away on each::labs. Grab the workflow below ⤵

[each::labs @eachlabs]

From AI agents to film studios, more and more of generative media now runs through each::labs. Our co-founder and CEO @eftalyurtsevenn sat down with @cnbceofficial’s Sıfırdan 1'e to talk about how we got here 🎙️

Here are just a few of the key points you’ll find in the episode:

→ AI agents are becoming some of our biggest users. Since we launched our MCP, they run the whole production flow through each::labs, so teams ship far more than before

→ New models come out every week. Our intelligence layer picks the best and most up-to-date one for every task, so teams don’t have to track every model themselves and they save up to 40%

→ Film and production studios now do a big part of their work on each::labs, from new camera angles to reworking scenes after the shoot

→ For retail, fashion or AI app teams running millions of generations a day, we post-train models built for exactly that job

There’s a lot more waiting for you in the full episode ⤵

https://t.co/GVPDHFOK4v

[each::labs @eachlabs]

Get every character right the first time: https://t.co/g5KhFiRHM7

[each::labs @eachlabs]

Want to recast EVERY character in a video?

Then stop wrestling with other models. They miss characters, give you plastic skin and break the small details.

Lockstep swapped the whole cast of this 19-second clip in one run for $0.90, while the others charged $5.60 and $6.40.

Right faces, 6x cheaper.

Bring your whole cast to Lockstep, without any bad surprises.

[each::labs @eachlabs]

London, are you ready for Nov 10? 🍻

@eachlabs and @vitpepper_stds are taking over London for Generative AI Builders Night.

Erim Şişman, Co-founder and CCO of Vitpepper Studios, and Korhan Erçin, Head of Business Development at each::labs, will talk about how teams are using generative AI today, from AI filmmaking to the numbers behind real projects.

Grab a drink and stay for the surprise at the end.

P.S. You’ll be one of the first people to see something new 👀

https://t.co/TtamDJxpcd

[each::labs @eachlabs]

Pick where you want to start:

Image to Video: https://t.co/46sgye0ft6
Text to Video: https://t.co/oXDwJxhnjM
Reference to Video: https://t.co/yi2v2R3eTt
Video Editing: https://t.co/CPcSxGxXwV
Video Extend: https://t.co/Qhyaqe7pqG
Video Upscale: https://t.co/u0lLORkRz4

[each::labs @eachlabs]

Gemini Omni 1.2 Flash is live on each::labs.

It makes clips up to 15 seconds and keeps characters, objects and scenes more consistent.

You can refine a clip turn by turn, swapping the background, subject or lighting without starting over.

Try it now ⤵

[each::labs @eachlabs]

https://t.co/P75gVWGNTy

[each::labs @eachlabs]

P-Video-2-Pro-Replace by @PrunaAI is now on each::labs.

Send a video and a photo, and get the same scene back with a new person in it.

→ Motion, camera and sound stay as filmed, with lips following the original audio
→ Replace one or two people, or just one person in a group
→ Clips up to 2 minutes, in up to 4K

Try it now ⤵