Decent4Apps

@zulali And the best level at that

@Decent4Apps mw2 undefeated

How im feeling with the new Sol

@rileybrown How do I know what they want.

Make something agents want.

@kylekane @rileybrown So true

@rileybrown Browser capabilities are endless

Very cool... You can now ask "ChatGPT Codex" to spin up many browsers tabs in the same session. Here's one simple example... the next example below is more useful 👇 🧵

@rileybrown @anshnanda Congrats

💯

@vercel @Lovable What about the backend

You can now deploy @Lovable apps to Vercel. vercel.com/changelog/you-…

@rork How is it

GPT 5.6 Sol is now live in Rork OpenAI's new flagship for coding and design. Second frontier model this week

@anshnanda Love it

New domain. Thoughts?

@NicolasLecocqHQ Well done

Day 1 of building in public. Goal: $10,000 MRR Current: $0 MRR I sold my last product to 500K users. Now I start over from zero, on purpose. Watch me rebuild it one day at a time.

@zulali Fascinating

i have been saying this. if the model thinks you you’re dumb, it becomes dumb for you. any hint of incompetence will let the agent lower its bar and begin doing the bare minimum of what it thinks you’ll understand. typos, dumb questions and ridiculous requests kill the models

@zulali Omg I miss this

@kylekane @theo Same issue

@theo Thank you for this the benchmarks don’t help me decide.

Over the last month, I burned over $200k in tokens with gpt-5.6-sol. I built a lot. Instead of just reacting to the benchmarks, news, tweets, etc, I went a different route. This video is an overview of all the cool shit I built using the new model. Proper review coming soon™

@kylekane @_MaxBlade Agreed

@_MaxBlade I’ve definitely had to correct it more than fable. But it seems to access tools a lot better. And they finally gave it design ability.

The truth about 5.6 sol after using it all day : The hype is overblown. Sort of. The benchmarks, and the commentary on X convinced me we were receiving AGI that runs at hyper speed, and is insanely cheap. in reality, 5.6 is built on the same spud pretraining as 5.5 this means its a nice bump, but not the opus to fable 5 LEAP in intelligence we recently experienced from anthropic. 5.6 is 2x times cheaper than fable on paper, and actually 3x cheaper when you look at actual task execution because of its token efficiency. BUT on swe bench where the models have to fix actual github bugs it falls behind fable pretty big. For vibecoders like myself this means I will be using 5.6 sol as a worker agent for Fable 5 to orchestrate alongside grok 4.5 I love this new era.

@kylekane @elonmusk @rileybrown Yep

@elonmusk @rileybrown I think you made it.

@kylekane More than most people can say.

got sober at 21 in a state where i knew zero people. no degree, moved to illinois after rehab with nothing. 14 years later i'm building apps at 4am before my CRE day job, running 11 AI agents off a Raspberry Pi in the basement, and shipping real products to real users. still no degree. MarketMaster day 11 with @shipordie_ and i already caught myself checking the deploy logs at 4:30am between sets at the gym. this is what build in public does to you.

@levelsio @levelsio building in public on X got my CMMS app its first real users. i was posting progress updates nobody was reading and then one facility manager DM'd me out of nowhere. shitposting your way to customers is a real strategy.

I made $1,247,389 in June 2026 Just Kidding I made $13.96 in June 2026 ProjectAdulting — $13.96

@kylekane Keep it up

> i have no cs degree, no bootcamp, no idea what i'm doing half the time. i just don't quit until the thing ships. > > turns out that's the only skill that ever mattered.

@zulali And they are?

i have 7 massive prompts saved and ready for when Fable/Mythos comes back online. i took for Fable for granted for the few days i had it. if it ever comes back online, every second is a gift by the state.

@kylekane where’d you go