aabhay

7 ms·

1.
by aabhay 5d ago
Note that ElevenLabs claims the right to train on any data you upload to the service, such as editing a song you made etc. > (d) License to Your Content and NILV. You hereby grant to ElevenLabs a license to use, reproduce, modify, adapt,
2.
by aabhay 5d ago
This probably corresponds to compute. ElevenLabs can burn compute for now, Suno can’t.
3.
by aabhay 5d ago
The music shows the post before legal got to put their hands on it lmao..
4.
by aabhay 6d ago
Formalizing proofs in Lean has gotten dramatically easier since the formalizations available in 2005. And Lean’s mathlib has done most of the underlying work so that you have its axioms and necessary lemmas baked in. You can think in terms
5.
by aabhay 6d ago
Cross shard transactions are “coming soon”. So, this is just like having multiple postgres clusters.
6.
by aabhay 7d ago
If the agent is able to “decide” when a loop should occur vs when an output token is produced, that effectively moves the CoT inside the architecture. While that’s not what is happening here, it’s clearly a plausible way we could see CoT di
7.
by aabhay 13d ago
This is with the caveat that OpenAI uses their own harness for this: > On ARC-AGI-3, GPT-6 Astra was run with our responses API harness , which changes two settings to better match real-world performance. The changes do not specifically
8.
by aabhay 19d ago
> There are some things that are unnameable Like what? Oh wait…
9.
by aabhay 24d ago
This concept of a free lunch was never true. In a competitive dynamic, speed and performance were always worth optimizing, comparing, and improving. One of the primary reasons for this is that computers operate in a vast range of orders of
10.
by aabhay 27d ago
In general, Claude family models write absolutely terrible comments, docs, and prose. It’s basically the number 1 thing that caused everyone on my team to migrate away from Anthropic models completely.
11.
by aabhay 1mo ago
With the Grok code incident in recent memory I have zero interest in trusting our company’s crown jewels with this service. Agents are here. Just self host and let the agents help cover your ass in network config, HA, whatever.
12.
by aabhay 1mo ago
A few years in between releases with a sane versioning system seems fine to me. If you don’t want new features don’t build with that new toolchain. It’s not javascript where you need to support all possible browsers.
13.
by aabhay 1mo ago
Everything but self hosted, is it?
14.
by aabhay 1mo ago
Genuinely curious what these constant changes are. In fact I don’t feel they’re moving fast enough to give us improvements to core.
15.
by aabhay 1mo ago
The strategy here is that the same services that Stripe offers for payments have corollaries in the LLM world. Security, user management, perhaps injection attack monitoring, etc. However I think they dearly overpaid for this as the core te
16.
by aabhay 1mo ago
It’s very clear from this article (and other product features and rumors) that Anthropic is teeing up for their next model release whose breakthrough feature will be the existence of capable agent collaboration. The irony behind this goal,
17.
by aabhay 1mo ago
The relevant paragraph about Meta’s commitment to open source is here, and the statement is significantly less confident than news is reporting > • Open source is a positive and important force for empowering people and preventing centra
18.
by aabhay 1mo ago
I have had a hunch for a while now that (in addition to these tools), Anthropic has actually leaned in to Claude's distinctive manner of writing since it makes the text more obviously AI generated and thus less susceptible to misuse.
19.
by aabhay 2mo ago
It makes it easier to write recursive async functions. It makes it easier for async functions to borrow rather than clone from their outer scope. All really awesome, non controversial and ergonomic things.
20.
by aabhay 2mo ago
Yes but the second order effect of this is that the cost of the tooling goes up since it is now the bottleneck, and therefore the shoemakers that survive do it off of technical complexity, branding, and regulatory capture.
21.
by aabhay 2mo ago
My main gripe here is the lack of transparency around the total experiment and construction. I doubt that they simply pointed their model at these ten specific problems alone and gave the model one shot; therefore the $2000 number could be
22.
by aabhay 2mo ago
Given that we were nowhere near this state even two years ago, I think it’s a question of velocity more so than just distance.
23.
by aabhay 2mo ago
Can anyone that works on this technology provide an honest assessment of where this technology actually stands? How much instrumentation is actually required, what the interaction quality is, how much trouble do humanoids have with in the w
24.
by aabhay 2mo ago
It’s funny how much of a “ChatGPT tell” that thin weight accent header is. Sol seems to place that on _everything_.
25.
by aabhay 2mo ago
While there is “law”, how much agency does an average European individual have in shaping those laws? This is a multi national body for which you can’t vote for its leader. You can’t vote on any of its resolutions or referendums.
26.
by aabhay 2mo ago
178kb?! What are you putting in there, a JVM?
27.
by aabhay 2mo ago
I’m curious what makes you believe that this distinction is precise or meaningful. What exactly is the difference?
28.
by aabhay 2mo ago
Un what? There’s several friendly clauses among these proposals including > 8. Any contributor who feels they cannot write in English without assistance, may write in their native language, and expect readers to use translation tools of
29.
by aabhay 2mo ago
I think the questionable term is not “syntactically likely” but “merely”. Syntactical likeness is a vast solution space that encompasses the work of a terrible developer and a genius developer. In fact this solution space is a gap wide enou
30.
by aabhay 2mo ago
Not necessarily. Drop semantics are just syntactic sugar, and can thus be aggressively inlined or auto vectorized by the compiler.

More ›