sohail

inference perf @togethercompute | ex @amazon @wendys
kernels, gpus, intelligence

[sohail @sohailmo]

@radbackwards would LOVE to learn about your design process/how you steered to what neo looks like today - pls write a blogpost

[sohail @sohailmo]

swe bench after openai called them out

[Scale Labs @ScaleAILabs]

We appreciate the community's feedback on SWE-Bench Pro.

Much of it maps to changes already underway in v1.1, which we've been building for a while. Keeping evals current with frontier models is hard, and we're always iterating.

SWE-Bench Pro Verified coming soon. 👀

[sohail @sohailmo]

@charles_irl i audibly cackled

[sohail @sohailmo]

sometimes i wish i was born 100 years later to have a museum to walk around with these things running

[The Humanoid Hub @TheHumanoidHub]

The engineers were allowed to cook.

[sohail @sohailmo]

admitting when you're wrong is a respectable sign of intelligence

[Elon Musk @elonmusk]

I was clearly wrong about Anthropic. They are obviously currently the leader in AI. No company has released a model as good as Mythos/Fable and they will undoubtedly have Mythos 2 ready soon.

And I would never cut them off in a way that hurt them badly, even as a competitor. That’s not my style.

Tesla open sourced its patents and we made the Supercharger network available to all competitors, even though we could have made it a walled garden.

SpaceX launches competing satellite systems with no increase in price or use of unfair terms.

Even my worst enemies can attack me on this platform.

…

[sohail @sohailmo]

this isn't to invalidate any effort the teams have put into organizing those artifacts or tools, moreso trying to find a more objective read on what is different here for my own learning

[sohail @sohailmo]

been thinking about this throughout the day and i'm now a bit skeptical as to how sol training luna is different from a pretraining engineer giving the tools needed to any model with a proper scaffold to iterate (similar to the plethora of auto research frameworks)

[jason @jxnlco]

Should AI be allowed to build itself?

I think RSI is gonna be the most consequential decision of the next decade.

Agree disagree?

[sohail @sohailmo]

or any /goal directive with the appropriate tooling/context/datastores and a box to run compute on

unless im missing some nuance, given sufficient local context on how pretraining/midtraining works, the model training another model still started with human organized artifacts

[sohail @sohailmo]

@thdxr ptu’s from togetherai

like rcu’s/wcu’s in aws but for inference

[sohail @sohailmo]

pair that with a cold/sparkling drink and you’ll be cruising

[sohail @sohailmo]

if you’re looking for an unconventional alternative to staring at a wall for an hour or touching grass to reset your screen-wired brain, go to goodwill and look through all the racks in your section

i promise you’ll be itching to go back to your computer after with ideas firing

[ClaudeDevs @ClaudeDevs]

We've reset 5-hour and weekly rate limits for all users.

[Tejal Patwardhan @tejalpatwardhan]

GPT-5.6 sol post-trained luna!

[sohail @sohailmo]

api to the physical world is such a good frame

[dar @radbackwards]

Alien Technology

[sohail @sohailmo]

using less tokens (cheaper) to get the same level of work done (if not better) is as much of a metric of intelligence as the end score is

comes back to the idea of being able to communicate an idea in as few words as possible as a marker of deep understanding

[sohail @sohailmo]

the 5.6 blogpost is v satisfying to read, lots of interactive bits to let you see the outputs yourself, v curious to test out the long context usecases - cautiously optimistic but scores showcase higher than opus rn (also abstract reasoning seems much better)

[sohail @sohailmo]

'looks like im finally allowed to talk about it...'

[sohail @sohailmo]

@matt_slotnick every post i see about sol sums to it being a diligent worker you can trust and fable being the insightful/intelligent lead, feels like it tbh

[sohail @sohailmo]

great read for anyone curious about agentic inference optimization

https://t.co/0hiprcDinn

[sohail @sohailmo]

cats will never be replaced by ai