fsd XV is good enough that every new disengagement is now just noise for their data loop (u dont need to specify disengagement reason now)
1
1
150
another yc startup trying to- wait a minute
Linux containers share a kernel. Docker Sandboxes don’t.
Docker Captain @yagcidev shows how your coding agent gets its own microVM, Linux kernel, and Docker Engine, with limits on what it can reach on your machine.
Read more: bit.ly/4eFXkDT
56
2,691
is this the first oss looptf? or first one that works well?
Some of you thought that Inkling 2 Small was too big at 594B parameters. We heard you: Inkling 2 Mini is a looped transformer with just 26B parameters, small enough to run on your laptop, powerful enough for agentic work with a score of 20.1 on AA Index.
thinkingmachines.ai/news/ink…
69
and space is fine because its united space of america so we are getting orbital compute at 2x fold every year in 3 years
62
can anyone translate this astra neuralese please
Replying to @mitsuhiko @badlogicgames
I can’t speak to all the reasons, but many things are more flexible if you shift to a model where the system prompt is not immutable at the top.
You can include more flexible environment and self knowledge context, for example.
1
15
2,701
umang retweeted
From Elon, I learned that most problems are actually totally fake if instead you decided to have courage
makes it very clear why they have to fucking build it themselves
“.. lenders have expressed concerns that if the group later decided to shoot the valuable chips that underpin their loans into space, it would prevent them from reclaiming and selling the semiconductors in the event of a default ..”
@FT $SPCX
ft.com/content/33b5643c-2957…
3
189
naming models random elements is sm kind of psychological warfare
like we wont even let u guess the fucking class of this model release go fuck urself
introducing Gemini 4 Neon: frontier-level intelligence with fast end-to-end task completion across complex software engineering, knowledge work, and science workflows
priced at $3 per million input tokens and $15 per million output tokens, with a 1 million-token context window
try it today via the Gemini API and in AI Studio ai.studio/build




















