shaped

Post training data for RL | prev 2x exits

[shaped @shaped]

@jjpcodes @shaunrein @vincent_koc real i might visit shenzhen in a week

[shaped @shaped]

@jjpcodes @YasmeenRoumie @vincent_koc oh wtf

[shaped @shaped]

@lewingtonpitsos Real

[shaped @shaped]

@vincent_koc @openclaw @steipete @nvidia @Microsoft lfg

[shaped @shaped]

Claude Opus 4.7 solves its first ProgramBench task 👀

After creating a 663-word CLAUDE.md we were able to increase solve percentage by 1.1 points on Claude Code + Claude Opus 4.7 across 10 ProgramBench, and actually solve one of them.

You can replicate our results here: https://t.co/TuTuKHZ6as

big thanks to @KLieret @jyangballin for releasing such a forward-thinking benchmark!

[shaped @shaped]

@jjpcodes Those mini shenjianbao look pretty fire

[emot-sun.gif industries @jjpcodes]

singapore eats part one; hainan chicken rice; insane buffet of crab, skewers, noodles, chicken; you name it, at Newton, food coma afterwards

[shaped @shaped]

@jjpcodes @msg @vincent_koc bro finally managed to look through all his photos on his camera

[shaped @shaped]

Hey Richard, congrats on the raise! We met at tony's poker a while back, and I came to your place for the games you hosted. Love Recursive's thesis, we have been looking a lot into auto research environments lately with a few other frontier labs since we believe this is the clearest path for whats next in AI developments, glad you see it the same way. Would love to send you some samples of what we've been working on in dms!

[shaped @shaped]

@jjpcodes @scheminglunatic Bruh 💀

[shaped @shaped]

@vincent_koc real

[Vincent Koc @vincent_koc]

For my eval-maxxing nerds out there, good friends of mine are running a series called "strange evals", you can benchmaxx now on anything. If in SF swing by! https://t.co/1LASlygFln

[shaped @shaped]

@morgantepell @cognition hi morgante

[shaped @shaped]

Happy to be working with meta for the past 7 month, and seeing the fruits of their labor.

Great release!

[AI at Meta @AIatMeta]

Introducing Muse Spark, the first in the Muse family of models developed by Meta Superintelligence Labs.

Muse Spark is a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration.

Muse Spark is available today at https://t.co/wHkMPH82ZH and the Meta AI app. We’re also making it available in private preview via API to select partners, and we hope to open-source future versions of the model.

Learn more: https://t.co/PloE9q5x96

[shaped @shaped]

@emilyinvc @Van0SS

[shaped @shaped]

@xdotli Hey sent you a dm

[shaped @shaped]

@FactoryAI Congrats on release! Great working w/ ya'll on it!

[Factory @FactoryAI]

No major benchmark is designed for COBOL, Fortran, or Assembly - the languages powering trillions in transactions and infrastructure that must be modernized or risk catastrophic failure.

We built Legacy-Bench to measure frontier agents on the code the world actually runs on.

[Ivan Bercovich @neversupervised]

https://t.co/buOPNmWeBe

[shaped @shaped]

@andrewjclare Well Apple themselves draw inspiration from Braun and Dieter Rams' philosophy, I guess it's all about stealing like an artist