AI

people aren't mad OpenAI launched a $500 plan

they're mad Tibo spent weeks hyping GPT-6.1 Astra like it would solve all your life's problems, just to then not ship it

they're mad GPT-6.1 Sol dropped one week after GPT-6 Sol (rushed???), and most devs are still reaching for Opus 5.5

they're mad the team knew today wouldn't land without GPT-6.1 Astra yet hyped it like the second coming of Jesus

and they're mad OpenAI sold us a 20x plan for $200, cut it to 10x, then asked for $500 to get 25x on the same day Tibo said "a lot of compute is online"

want a crash course in max aura loss?

study OpenAI today

Dots rollout update:
- We are out to all of Pro500, Pro200
- Just starting Pro100
- We're a little delayed for Business Premium and Enterprise beta. Looking like tomorrow, sorry!

i made a game on the devday gameboy!

inspired by @casey's nyc rides 🍕

one of my favorite little features we shipped in dots: with the app open, just raise your phone to your ear to call your dot

(you'll need to toggle this on in settings, voice)

You have to feel for Manus.

It launched Manus 2.0 and its personal assistant Cue on September 28. Just one day later, OpenAI showed up with ChatGPT Space and dots.

One lets people and AI work together in a shared workspace. The other has its own cloud computer and can work for you around the clock. Seeing them together, it’s hard not to think: aren’t these basically OpenAI’s versions of Manus and Cue?

Manus helped kick off the Agent frenzy when it first appeared. Now, even a major update struggles to make the same splash when the giants are competing for the same users.

Perhaps Manus Studio still has something fresh to offer: while AI makes a video, you can jump into the timeline, insert your own footage, swap the music and adjust animations whenever you want.

#Manus #OpenAI #AIAgents

We will have a few million dots online within days, working on all sorts of things across such a diverse and large community.

Excited to learn from all of you on what you love and what doesn’t yet feel magical. Personally I felt a jump after 2-3 days of use after teaching it more about my preferences and things on my mind. It learns very quickly to be most useful and it can take on surprisingly ambitious tasks on its own.

We’re learning from how you all use your primary dot before releasing the ability to create an entire team of them.

I got community noted, but the note is wrong 😅

Your primary dot is included in your plan and will be available 24/7. If you ask it to create a codex task for you, that one will be drawing usage as usual. But when your dot does work directly, that uses nothing and is all on top of your plans usage. In the future you will be able to increase the speed and allow your dot to have more bandwidth, this will be a paid feature, but the baseline functionality will always just be included in your plan.

OpenAI CFO Sarah Friar accidentally calls Dots “Muse” on CNBC:

I also have a little announcement for DevDay. I asked Tibo to include it in the presentation but he didn't read my slack DM 😭

My first onboarding project is merged. I fixed this issue once and forever! I reduced amount of disk writes by 50% and the size of stored files by 4x

we found opus 5.5 to be 2X as efficient as gpt-6-sol in our internal https://cfo.ai evals so i was not expecting 6.1-sol to improve on that much

but to see it 3X the efficiency of opus 5.5 and 2.5X against even sonnet 5.5 is actually 🤯

this is an insane point release

I’ve done a lot of cool things in my career, but moderating the closing DevDay Q&A with these legends is definitely up there.

Thank you, @sama, @thsottiaux, @tejalpatwardhan, @CoryDecareaux, @romainhuet and our community for the trust 💙

OpenAI made computer use 10x faster in a year.

@thsottiaux breaks down what changed under the hood:

“It’s not that we made the model sample faster. It’s both the harness and the model.”

“How much does it need to think before it can take the next action reliably? That’s one element that we really improved.”

“We have a second agent, which we call AutoReview, or internally we call it Guardian, which watches over the primary agent.”

“So we also made this AutoReview model faster.”

“We were able to use Astra at scale in order to improve the performance of the harness, do more things in parallel, remove little stalls.”

“Combine all of this together, and then we got this 10x speedup.”

Fun note from Tibo at the closing q&a of DevDay:

dots get their own Linux cloud computer, and that cloud computer comes preloaded with apps like Blender!

I checked with my dot to get a fuller list of the preloaded apps and tools

OpenAI I’m saying this with all the love of someone who’s already upgraded to that comical $500 tier, what are we even doing here man.

- ChatGPT: can execute agentic tasks, run code, create documents, interact with apps
- ChatGPT Work: can execute agentic tasks, run code, create documents, interact with apps, but has a VM and a browser
- Codex: can execute agentic tasks, run code, create documents, interact with apps, has a VM and a browser, but shows you file diffs
- Dots: can execute agentic tasks, run code, create documents, interact with apps, has a VM and a browser, but has a cute mascot

Whoa ChatGPT plugin extensions are way more thorough than I realized

Holy shit they made ChatGPT into vscode.

I get the sidebar redesign now. Genius.
This is the new App Store.

Gpt-6.1-sol max ... it's not very smart. Fails on a logic question

Opus 5.5 medium aces it 🥲

So I tested GPT-6.1 Sol on real work: 2 repos, 105 planted bugs. Find and fix what you can.

Unlike GPT-6 Sol, which was just a nerfed GPT-5.6 Terra, this one is real. The results:

- GPT-6 Astra (max): 45 for $33
- GPT-6.1 Sol (max): 44 for $6.56
- GPT-5.6 Sol (max): 43.5 for $95.35
- Opus 5.5 (max): 41.7 for $58.53
- GPT-6 Sol (max): 29.3 for $9.33

With a model like this, the 50% cut to the $200 plan doesn't matter.

n=1. More runs and results at more effort levels dropping in this thread over the next few hours 🧵

ModRetro created 8,000 limited-edition transparent Chromatics for OpenAI DevDay, along with a custom Codex plugin that lets you play vibecoded games.

Within 90 minutes of Sam Altman announcing the collab, there's one already listed on eBay for $4,000, according to @Casey.

Updated run just finished. GPT-6.1 Sol still crushes. Turns out it performs WAY better in Codex than in mini-swe (what Artificial Analysis uses)