Lovable

before Lovable starts building, it sketches out what your app will look like, super cool to see before the actual building begins.

i built a meme generator that creates memes based on my prompts. It picks the image and writes the copy

You can now chat with Lovable about your relationship problems, just don't forget to build.

Million dollar product idea: chat with your Google Calendar.

Powered by GPT-6 Astra.

You can now chat with Lovable about anything.

Think out loud (literally, with voice support). Chat has deep context on your Lovable apps and connected tools, and it’s free.

A sign from the universe 🫶

Doomscrolling was our #1 social monitoring tool at Lovable, so @thisisyasamin and I built Capybara. It listens to what users say about us, in real time, for humans and agents.

Now it's one data layer powering social engagement, event monitoring and malicious site screening.

Lovable is now in the @claudeai Startup Stack.

Claude Startups members get 4 months off a Lovable Business plan. Learn more: claude.com/resources/arti…

Over the last 30 days, Lovable Realtime sent 1.11 TB of application payload. If we'd sent every result as a full snapshot, we estimate it would have been around 40 TB.

Three optimizations account for the difference:

1. Don't send what didn't change.
A change in the database doesn't always change what a subscriber sees. Maybe it touched a field they can't read, or only bumped a timestamp. So we recompute their authorized view, fingerprint it, and compare. 88.4% of the time nothing had changed, so we sent nothing.

2. Send a patch, not the whole thing.
When the result did change, we send a patch if it's smaller than a full snapshot. New subscriptions still got the full snapshot. Patches cut the payload volume of changed updates by 94.5%.

3. Don't resend what the client already has.
When a client reconnects, it tells the server the fingerprint of the state it's holding. The server rechecks authorization and loads current data. If the fingerprints still match, it replies with a tiny "resumed" frame and the client keeps what it had. On top of the first two, this saved another ~50%.

All in, that's about a 97% reduction. 💪

I built a small simulator so you can watch it happen. Change a field, revoke a permission, or trigger a deploy, and see exactly what goes over the wire:

lovable-realtime-simulator.lovable.app

Okay, I promise this is the last teaser 😅 Blog post coming very soon ⚡️🚀

We got Lovable OGs on stage to share stories from the very early days.

We built an internal tool so your feedback doesn’t get lost in the feed.

It’s a custom social listening platform + agent that finds your comments, questions, and Lovable builds across channels.

So keep tagging us, we're here on the other side!

You can now add live voice conversations, powered by @OpenAI's GPT-Live, to your Lovable apps.

@mgfeller built one where you can talk to Rocky from Project Hail Mary.

Amaze amaze amaze.

Lovable + @Microsoft: build tools that reduce manual work with your Microsoft data.

Connectors are available today. Publishing apps into your company’s Microsoft environment through Copilot Managed Runtime is rolling out in public preview.

Every small business has a tool they wish existed.
Now they can ask Muse to build it with Lovable.

We did it. And it's free. Chat with Lovable now.

Freelancers and agencies who build with Lovable: @contra just handed you free Lovable Pro and a chance to compete for a $25k prize pool.

The brief: create a hypothetical appointment-based business.

1st place: $8,000. 10 total winners. Go build.

You can now add AI features, powered by the latest @claudeai models, to your Lovable apps.

Here's a favorite app from our team that uses Opus’ visual identification and deep reasoning skills + Sonnet’s fast web search, retrieval and synthesis.

2 new models powering Lovable today: Claude Opus 5.5 and GPT-6 Sol.

You don't need to think about which to use. Lovable combines the best models to get the result you want.

One more thing… Lovable now builds with GPT-6 Sol.

On our 0-to-1 building benchmark it scores 6 to 12% higher than GPT-5.6 Sol at every effort level. The biggest gains are in following the brief, 10 to 15% higher on intent alignment, and in design fundamentals, 8 to 10% higher.

Starting today, builds in Lovable are faster, without any drop in quality.

Lovable now builds with Opus 5.5.

Same quality as Opus 5 on 0-to-1 building and iterative code fixing, at up to 48% fewer steps on existing codebases. It's also more disciplined about checking its own work, scoring 4-6% higher in self-verification testing at up to 57% fewer steps to get there.

Learn more: lovable.dev/blog/opus-5-5-…

Applications for SheBuilds Season 4 are open.

72 hours. One idea. A real product at the end of it.

Apply now: shebuilds.lovable.app