For the past few months I've had sub usage anxiety Not having enough usage to do what I need - I even had to take days off to ration it some weeks Now with opus 5.5 I'm throwing the kitchen sink at it Now I have anxiety about not making full use of the sub smh
3
3
118
these actually look great
Sawyer Merritt
2
97
Fable 5.1, I hardly knew her
1
1
107
Feels good to have something worth using the other 50% of the usage limit on!
23
Wow Opus is back They fixed our boy
Claude Opus 5.5 is now available in Cursor! It's the new top model on CursorBench at 57.8% (Max) and costs 40% less per task than Opus 5.
1
2
161
Claude banked resets And a potentially “usuable opus” model 👀
Replying to @claudeai
One more thing: we’re increasing five-hour usage limits on Pro, Max, and Team plans. We’re also providing subscription users a rate limit reset, which you can save and use whenever you choose.
164
Current stack 2x Claude Max for real work with fable 1x Codex Max on a Mac mini as a literal employee 1x Cursor pro for a lil grok bot , lot of automations breaking from data centre ips this week 1x Hermes for ??? 1x Instinct, literally only finds things to do LDN Cooked
3
3
212
Not even 24hr 🚬 🚬🚬
1
1
121
Matthew Fox retweeted
Replying to @jamesm
If you ask your mom "is the sky blue" she says "yes" immediately, that's jev. If you ask your dad "why is the sky blue" he talks for ten minutes, that's an LLM.
47
119
1,738
53,296
Jev could do this in minutes, no seconds
How they texture a wall
3
352
I added jev to a mobile game to help me test multiplayer because default bots are fucking dumb It’s kicking my ass 😭
1
1
126
Ok i finally won Jev 7 - me 1
32
This could unlock a lot Faster cheaper intelligent systems that don’t get lost in their own sauce There’s a lot of opportunity in pairing this kind of system with llms **If it’s good**
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x cheaper (w/ output tokens free) • Frontier composable intelligence optimized for decisions AFAICT the shortest path to AI-based economic revolution
1
138
This is actually legit this time Blender, unreal engine, play testing - Astra genuinely can do it all It even sources and watches references to improve its own animations Look at them traps pop on the kick
1
1
4
508
That said I'm 3 banked resets deep today 💀
Astra + Computer use on Mac is pretty much opus 4.5 for coding moment but for everything else you do at a computer 1 day into using it on a Mac and it just fckn rips
1
1
111
I miss when @warpdotdev was incredible Was excited about the opensourcing and the concept of 0z and everything But its diabolically bad, slow and laggy when it used to be truely blazingly fast Might have to fork and rip out all the bs or just drop it
2
129
Astra with computer is so good I had to get a cloud mac Computer use on windows is abysmal, you can't even use your computer while its running unfortunately I still main windows 100x experience on mac
1
116
so all you need to be in the <1% is $20 to use @bot or @NousResearch nice
You do realize that if you’re coding on your phone with agents or bots, know how to use computer use with a sandbox and built your own harness you’re in the less than 1%
2
147
viva la astra
1
99