Introducing Merge Gateway - Build Your Own Router.
You're three sprints into your coding assistant.
You pick the most hyped model, integrate, test, deploy.
A month later, a new model drops.
Now you re-test, re-integrate, re-deploy.
Your product didn't change, but the benchmark did.
That's how most AI teams operate.
Chasing a "best" defined by people who've never seen their product.
There is no best model.
There's only the right one for your product, users, and use-cases.
Build Your Own Router runs on your definition of good.
Pick your benchmarks, weigh them, add your own evals.
@merge_api routes every request to your winner.
👉$100 in credits to the first 200 people that comment
https://t.co/VLbDYycQRY
We live in a golden age of biology. So why are people still dying from disease?
Because discovery and development move slower than they should.
Today, we’re partnering with Incyte to change that.
Kosmos is now the first agent that can compress months of drug development into weeks, from the earliest stages of scientific discovery through to FDA approval. @Incyte will be the first company to deploy it across their pipeline.
Work that used to take a team of scientists months now happens in weeks.
Patients can't wait, and neither can we.
Every AI workspace lets you ask questions.
Almost none let you bring your own knowledge into the conversation in a structured way.
Sources changed that.
A knowledge library you control. Upload files, documents, and references OmniAgent treats them as the primary context for every answer.
Stop pasting. Start querying your own library.
Live in Omni Agent: https://t.co/3r7nO2TGDf
Introducing Buzzy’s Baseball Cam Filter.
Want to follow the Korean Baseball trend and be caught on camera?
No more wondering how it’s done—it's all in one click with Buzzy.
Your viral moment starts here: https://t.co/SPciw3dybi
Introducing the Cline SDK. We rebuilt the Cline harness for our extension and CLI from scratch using all the lessons learned since creating one of the world's first coding agents in 2024, and are open sourcing it for others to build with today.
npm i @cline/sdk
🧵
We got an 8-figure acquisition offer 2 days after launch.
We said no, because the problem we're solving is worth way more than that.
It’s 2026, but teams are only getting lonelier, and context is still the problem.
The issue isn’t intelligence. Your team has plenty of that.
It’s shared memory and context, the thing that makes 10 A-players feel like 1.
That’s what we’ve solved with @playdotfast, while making work more fun.
We're killing traditional SaaS, and believe you me, we're leaving no holds barred.
AI is not going to replace human beings,” said @mcannonbrookes speaking to @l2k on @wandb’s podcast.
And it definitely stayed with me.
AI & agents are only going to enable us to do more, build more, and help humanity realize its full potential.
#TLVCPartner
Introducing SubQ - a major breakthrough in LLM intelligence.
It is the first model built on a fully sub-quadratic sparse-attention architecture (SSA),
And the first frontier model with a 12 million token context window which is:
- 52x faster than FlashAttention at 1MM tokens
- Less than 5% the cost of Opus
Transformer-based LLMs waste compute by processing every possible relationship between words (standard attention).
Only a small fraction actually matter.
@subquadratic finds and focuses only on the ones that do.
That's nearly 1,000x less compute and a new way for LLMs to scale.
Been dogfooding Sai with Claude Code lately and the part that's actually stuck for us isn't just the responsiveness. It's that you can get work done on your phone. We have teammates improving our app from iMessage daily now.
This all started last week, when a teammate spotted a UI bug (black text on a black button). Normally that's a ticket → repro doc → Slack thread → fix-tomorrow situation. Instead Sai reproduced the bug, took screenshots, and handed it to Claude Code. Put up a PR before someone even read the bug report.
Whole thing started on our teammate's phone. On the go, laptop never opened, all in our VMs.
In preview now. Try it at https://t.co/WLHgZN9UJG with invite code: SAI_CLAUDECODE
The "look what AI did" reels skip the part that matters: how it was directed.
Vishal Balsara, our Creative Director, built a 7-min Hachiko short in 3 days on Agent One and recorded the full 41-minute tutorial. Context, treatment, shot-by-shot.
Film below. Full tutorial in the next tweet👇
There's a gap in the AI filmmaking workflow that nobody talks about: you fix one thing, two things break, you start over, and repeat till midnight until you've lost the last drop of creative juice.
Here's how to make sure that never happens again.
Who's next on your list? Drop your tribute suggestions in the comments.
With Agent One by invideo, the way you think about films,
is now the way you make them.
Stop Prompting
Start Directing
Just tell Agent One what you want and it does the rest.
Introducing Smart Shot - Powered by GPT Image-2 and Seedance 2.0 ✦
One sentence. A complete cinematic scene.
→ Full Shot Plan (characters, environment, storyboard, camera)
→ Multi-cut video (wide, close-up, cutaways)
→ Consistent characters across every frame
→ 10-20 seconds, production-ready
If your team is still manually adapting ads for different markets in 2026 - you don't have a creative problem, you have a (very fixable) tooling and workflow problem. Here's what intelligently creating, editing, and adapting films and ads looks like👇
Kling 3.0 is live on OpenArt, the first AI model with native 4K video. No upscaling.
-Production-ready visuals from a single prompt
-Built for ads, film, and broadcast
-Consistent subjects, clean text, controlled lighting, all at native resolution.