Nice. But my MacBook Pro RAM can barely survive with a few Claude Code and nextjs builds today so…
We are proud to introduce our new product, VMPal.com
It is a Virtual Machine specifically designed for Apple Silicon and AI Agents. This is our very first release, and it fully supports macOS, Windows, and Linux with GPU acceleration.
- Liquid Glass,
- GPU Accelerator.
- MCP for AI Agent on your Mac or inside the VM.
Andrea Falzetti retweeted
Dev Day announcement overview, as short as reasonably possible.
- Dots: personal assistant (similar to GrokBot, Muse etc)
- $500/month plan including Ultrafast
- GPT-6.1 Sol: really smart and cheap, way better than 6 was
- Decisions API: Jev killer with vision capabilities
- ultrafast is real (300tps on Astra). It’s way expensive though.
- sign in with ChatGPT: use your sub and inference in 3rd party apps
- extensions in ChatGPT can now be full apps
- 1 banked reset for all
I think that’s most of it!
I see a lot of folks sharing game prototypes made with Opus / Astra.
Has anyone actually shipped any game to the stores yet?!
I have also been grinding and I am kinda close to release the beta of my own game but the more I work on it the more I want to add things before launch.
This is great developer experience. It’s not clear how much how this Codex already does, I think it deserves more transparency on both sides.
Andrea Falzetti retweeted
Switching effort mid-session on Opus 5.5 doesn't break your prompt cache btw! Just make sure you're on Claude Code v2.1.280+
Am I the only one that thinks Jev is not really useful in a real word scenario?
Introducing 𝚒𝚜𝙴𝚟𝚎𝚗 / 𝚒𝚜𝙾𝚍𝚍, but with @typesafeai Jev + @aisdk
Now available on NPM: 𝚗𝚙𝚖 𝚒 𝚓𝚎𝚟-𝚎𝚟𝚎𝚗-𝚘𝚍𝚍
I have been using deepseek 4.1 flash Max for a few days now and I can I trust it for well scoped tasks, however planning architecture or code reviews you sill need Sol xhigh at least.
Same here.
DeepSeek v4.1 Flash might actually be good enough for the majority of my work.
I’m spending $300/month across two Codex accounts and still constantly worrying about limits.
Meanwhile, I’ve had 5+ agents running on DeepSeek all day and I’ve used 15% of my weekly allowance.
$10/month (@opencode) vs. $300/month.
At some point, “good enough and available” beats “better, but you can’t use it.”
It seems the incident only lasted 3 minutes and is resolved indeed - but you have to kill dsh web and start your conversations or they got stuck.
It was fixed quickly and it’s working great so far.
@thsottiaux their cache hits is 99%, amazing performances and predictable costs. What are you guys waiting for?
The x20 is still not available, how are we supposed to use Codex if you guys don’t provide a proper plan?
Decided to give it a try. Let’s see - will report back.
I am trying to use DS daily on simpler tasks and save on Codex/Claude x20 accounts because they are draining too fast.
Using OpenCode so far, but if it works better on dsh it’s worth a shot.
Andrea Falzetti retweeted
first "factory machine" mechanic: looms
add yarn and it will afk-produce cloth for you
Andrea Falzetti retweeted
Give us a slow mode for Codex.
Something like /slow where it runs at half the speed but uses roughly half the usage. It would put less strain on OpenAI’s servers and maybe even make the 20x plan possible, right?
It’d be perfect for /goals too, just send a task before going to sleep and let it work overnight.
I’d happily trade speed for more usage @thsottiaux
🤖 Made with AI
