SPUD SMASH made with fable 5.2 on medium
play as a ๐ฅ and destroy as many blue chairs as you can
punch them, headbutt them or set them on fire
i got 3 fable resets on one of my claude accounts
i thought it was a bug at first but nope
if your account is more than 12 months old you get 3 fable resets
6-12 months = 2 fable resets
under 6 months = 1 fable reset
this is for max and ultra, the age counts from the day you first paid
you have until march 8 to use them
Show more
We wanted to do more for the people whoโve been with us the longest. So with Fable 5.2, the longer youโve been subscribed, the more Fable resets you get.
no way, my fable 5.1 is being routed to fable 5.2
running some sick prompts, will report back ๐ซก
anthropic: here is fable 5.2, it's the best model in the world + fable resets
openai: dots can connect to multiple computers ๐คช
You can now connect multiple local computers to the same dot and choose which one it uses for a task.
It can build and test your app on a dedicated machine while you keep coding on your laptop.
Show more
heads up, the new fable reset is not a normal weekly reset
it only gives back the usage you spent on fable, everything you used on other models stays where it is
on one of my accounts my weekly is at 99% and only 1% of that is fable, so if i press it now i get back... 1%
use it right after your weekly resets: run only fable until it hits its cap (half your weekly), then reset and you're back at 0%
time to fablemaxx folks, happy times ๐
Show more
Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges.
Theyโre capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.
Show more
holy shit, fable 5.2 just dropped and it's a beast of a model
on high it scores about the same as opus 5.5 on max in anthropic's coding benchmark, for roughly HALF the cost per task
on xhigh it gets the highest frontiercode score of any released model: 49.1% vs opus 5.5's 45.5% ๐ฎ
and it's cheaper too, $8/$40 down from $10/$50
anthropic is on a roll recently
Show more
Fable 5.2 and Mythos 5.2 are a step above Opus 5.5 in almost all the capabilities we evaluated.
Testers have found that they communicate more clearly than past Fable and Mythos models. They generate outputs up to 30% faster than Fable 5.1 and Mythos 5.1.
Show more
ferrofluid toggle made with opus 5.5
no three.js, no libraries, just one html file, even the sounds are generated in the browser
agents in kylon just crossed 5 million transactions, about 50,000 hours of manual work
kylon puts your team and your claude code and codex in one room, so anyone can @ an agent and the request and the result stay in one thread
and it's more than code: research, marketing, sales and ops, with 3,000+ tools like patent databases and meta ads
anything that changes something outside the room comes back as a draft for a person to sign off
Show more
Agents on Kylon have now completed over 5 million transactions. That's about 50,000 hours of manual work.
Bring your team, Claude Code, and Codex into one shared workspace where agents can handle research, marketing, sales, and operations alongside your people.
With access to 3,000+ tools, from patent databases to Meta Ads, your agents can work together across your business.
Comment with your use case for your first month free.
Show more
@daniel_carvalho Really like Russel Brand. Will check it out. Surprisingly deep thinker.
1. opus 5.5 did better
2. this is not gpt sol 6.2
There is NO WAY GPT-6.2 Sol just beat Opus 5.5.
After some back-and-forth on, I was finally able to get my hands on the routed GPT-6.2 Sol through a friend.
I tested it against Opus 5.5 using the exact same prompt, both at their highest reasoning settings, and the results genuinely surprised me.
Here's how they compared:
- GPT-6.2 Sol: Took around 20 minutes and cost $9.
- Opus 5.5: Took 1 hour and cost $20.
Sol nailed the details really well and produced something you'd expect to see in a low-budget animated movie.
Opus 5.5 focused more on the overall environment, and it looked really good too. But I have to be honest, Sol did better here.
Which one do you think won this comparison?
Show more
the only real "i'm being routed" post i found
Holy shit guys Iโm being routed to GPT-7-Sol. Hereโs a prompt you can try to find out:
nvidia ceo jensen huang emailed the whole company: "i was unfaithful to my wife"
the nyt says the woman is alice huang, who ran megaspeed, a singapore company that bought over $4.8b of nvidia chips in the last two years
us intelligence concluded she was spying for china
nvidia's board is investigating, says it has no reason to think any secrets leaked, and jensen stays ceo
i can't believe this is real
Show more
Breaking News: Nvidia opened an investigation into Jensen Huang, its CEO, after he acknowledged to the companyโs board that he had been intimate with a business executive who U.S. intelligence determined to be a Chinese spy.
Show more
opus 5.5 ran 1,062 chats for me this month and i only started 328 of them
polaris, my orchestrator agent, started 376
my other agents started 358
gpt 6.1 sol: i started 9, my agents started 674, mostly reviewing code
sonnet 5.5 mostly works as a subagent, 875 runs, mostly research and code
a month ago i started 138 chats a week by hand, last week it was 31
i just talk to polaris and it runs the rest
Show more
The gap between an average engineer and the best engineers was never quite 10x, but it was meaningful.
AI has made the gap at least 100x.
i'm all for copium and hopium
but this doesn't prove it's 6.2 sol
p!nk was announced as host in april, before 6.1 sol's knowledge cutoff
and "i recognize it, less sure about details" is vague, something a model could say without actually knowing
i would love to be proven wrong though
Show more
I can confirm that GPT-6.1 Sol has already routed me to the more current model (GPT-6.2 Sol?). Two testings:
-(Simpson) 742 Evergreen Terrace as an interactive Three.js diorama
-Pelican
Show more
people are saying opus 5.5 got nerfed on nov 3, so i checked my own claude code logs
67k opus messages, before vs after:
tool calls per message: 1.06 โ 1.06
failed tool calls: 1.9% โ 1.8%
me getting mad at it: 15.7 โ 14.2 times per 100 messages
not seeing a nerf on my account
Show more
how long since openai and anthropic released their biggest models:
claude fable 5.1: 68 days ago
gpt-6 astra: 66 days ago
claude opus 5.5: 47 days ago
claude sonnet 5.5: 41 days ago
gpt-6.1 sol: 40 days ago
gpt-6.1 terra: 13 days ago
i'm starving for a new model, it's time
Show more
yeah grok bot is nice, but
> needs local computer use asap (please @poteto)
today is day 1
we better get something that is a CLEAR improvement or i will need another banked reset sir
Over the next 28 days, each day weโll either ship one thing that is a clear improvement and relevant for most codex/work users or ship a full reset. Let the improvements begin.
could have done this in a normal codex thread and probably faster ๐คท๐ปโโ๏ธ
but sure
my dot has saved me over $2,000
