Bruno Lourenço

Morphing animations should not be hard. I'm super excited to announce my #react-morph 🎉 Just define the From and To states with CSS, the rest is magic! github.com/brunnolou/reac…

@thsottiaux @adonis_singh 🥔?

@adonis_singh Uhm

it’ll probably be months before we use a model of this level of capability

@RoyalCities Amazing! But now, for the rest of the mortals, could someone train a small model like Gemma to "translate" normal prompts into your training tags?

After months of work, today I’m releasing Foundation-1. A SOTA text-to-sample model built specifically for music production workflows. It may also be the most advanced AI sample generator currently available - open or closed. • ~7 GB VRAM • Entirely local • 100% free 😁

@ZoldenGames Who would guess that stirring is a good way of mixing? 😅 Cool demo, what tech?

Mixing fluid by rotating wheels.

@sama @grok context please

We love working with NVIDIA and they make the best AI chips in the world. We hope to be a gigantic customer for a very long time. I don't get where all this insanity is coming from.

@ImLunaHey @kristianfreeman @openclaw @moltbook I don’t want to increase the surface for prompt engineering

@kristianfreeman @openclaw @moltbook curious why not?

@aisdk ... baby, don't hurt me. don't hurt me no more

Use Gemini 3 Flash with the AI SDK and Vercel AI Gateway.

@aayushkapoor_ @aisdk Thanks! 🙌

@brunnolou @aisdk yes you can check out the an example here: github.com/vercel/ai/blob… specifically the 'libs' imports

Use OpenAI's apply patch tool to let GPT-5.2 create, update, and delete files using structured diffs.

@_akhaliq Is this just an upscaler?

Apple just released Sharp Sharp Monocular View Synthesis in Less Than a Second huggingface.co/apple/Sharp

@colmtuite That’s exactly what I was hoping for! 🎉 Congrats @colmtuite 👏

shadcn 🤝 Base UI A first class integration. Same shadcn abstraction, rebuilt for Base UI components.

@IOivm Is it possible to simulate gravity pulling drops like graffiti?

5 ink effects for a digital brush in p5.js • Edge Erosion — soft, absorbed edges • Squeeze & Spots — bright center, darkened edges • Dry Edge — marker-like outlines • Salt Texture — chaotic, organic noise • Wet Bleed — watercolor bleeding

@colmtuite I'd hope v1 gets wide acceptance! Do you think it would be possible to be officially supported by shadcn ui? Any thoughts @shadcn?

We just shipped a v1 release candidate. All going well, stable v1 should be next week.

@rauchg @forgebitz Its it free? What's the limit?

@forgebitz v0nanobananapro.vercel.app

i am going to sound like a boomer but why is it so hard to run nano banana? like i google for it, 20 other sites, finally find gemini turn on "nano banana" but it's not the latest one get some garbage images have to figure out how to run the "pro" version? turns out it can't edit people anymore 90% of use cases for image enhancement/edits gone..

@jjlin_dev @nicoalbanese10 But, if I’m not mistaken, only use the thinking tool on non-thinking models, otherwise the gains are minimal or none

@nicoalbanese10 they also recommend having a "think" tool to just tell the llm to take a minute and think about all the steps it needs to work on

Find it fascinating that Claude Code has had so much success in the no-op TodoWrite tool approach for planning Feels very fragile to have to regenerate the entire plan at each step just to update it - but their models use it so well

@bradkowalk This is a great UI pattern! Wish you success with your product. And, yes it won't be very pleasant to have ads on autocomplete, but I guess this is the world we live in. Nice idea. But you know this pattern will be immediately copied, right? Typescript implementation, anyone??

I’m very excited to announce AI Autocomplete AI Autocomplete is a breakthrough patented technology that supercharges natural language input = Unlocking 10x faster search and commerce, advertising, and powerful augmented reality. Available for use as an SDK. This solves a fundamental problem of today’s chat interfaces – They are good at single step request, but any multi-step action (like booking a flight or purchasing goods) quickly becomes a back and forth 'game of 10 questions'. While working on our own assistant, we realized the core driver of this problem is a human one: People don’t know everything they need to say upfront, for every action you could do on the internet. And to solve it, we would need to think about how to marry design and technology in a new way. AI Autocomplete solves this, and can now plug into chat interfaces to guide you in real-time, with everything you would need to say, upfront. So you can do anything you can do on the internet in one shot. No more back and forth. As a result, this unlocks multiple AI breakthroughs: 1. 10x faster search and commerce 2. Smarter (and far lower cost) media generation 3. Natural language advertising 4. Powerful, lightweight Augmented Reality From here, we’ll be using this for Hero, but we also want others to use it too since this is an industry-wide problem. So if you have a product that could benefit from AI Autocomplete and want to work with us, reach out below! Shoutout to @seunglee1b who helped think of and patent this nearly 3 years ago! And shoutout to the entire @hero_assistant team that keeps innovating on the next generation of AI products

@NathanFlurry @rivet_dev Thank you, it was helpful

Hey! There's some overlap, but most users use Rivet Actors for more advanced use cases than workflows that require in-memory state. We also don't have a first-party durable execution API, but we've been testing one internally for a while. Use case is still different. e.g. We support realtime (including running TLDraw, Yjs, local-first sync), multiplayer, per-tenant databases, background jobs, etc.

@NathanFlurry is Vercel’s “use worflow” a direct competitor of @rivet_dev? If not, what’s the USP?

@tldraw This brain effect is what makes Apple Liquid Glass look good on promo videos or while scrolling, but as soon it is still, our brain merge the background with the foreground and make it hard to read 💡

if you pause this at any moment the tldraw disappears

@mattpocockuk @_cantemizyurek More than being the unit test for LLM, have you considered having tooling for manual labeling and human evaluation? Have you watched this episode? spotify.link/raSYGPcJEXb

My dream for Evalite is to be THE evals tool for the AI SDK. Scheming with @_cantemizyurek to make this happen.

@lgrammel @TobiasBueschel Does it work with Vercel's AI-Gateway?

Shipped for AI SDK 5 and 6 beta: Anthropic Agent Skills Thanks @TobiasBueschel for the PR!