Hacker News: swingboy

New comment by swingboy in "US imposes sanctions on ICC hours after former judge wins Nobel Peace Prize"

swingboy — Fri, 09 Oct 2026 17:38:47 +0000

The Israeli lobby poured a lot of money into his campaign after 10-7 and the entire reason the current administration even has beef with the ICC is because they issued arrest warrants for Netanyahu.

New comment by swingboy in "Decisions API is in public beta"

swingboy — Wed, 07 Oct 2026 10:04:44 +0000

I think the hardest part would be coming up with the instructions/inputs to send for every non-trivial action?

New comment by swingboy in "Powerless F1 drivers frustrated by Bahrain F1 software glitch"

swingboy — Mon, 05 Oct 2026 11:12:44 +0000

This must be related to how Toto turns off George Russell’s car if it’s looking like he’s still a potential threat to the golden child, Kimi. Kidding. Kind of.

Signed, a bitter George Russell fan.

New comment by swingboy in "Getting the most out of Opus 5.5 in Claude and Claude Code"

swingboy — Sat, 03 Oct 2026 23:25:56 +0000

Just create two $20 plans and swap between them when you hit limits.

New comment by swingboy in "[dead]"

swingboy — Thu, 01 Oct 2026 20:43:30 +0000

Manyfold's Agents.md Is Cringe*

New comment by swingboy in "Clef: Open-source decision models, and new RL fine-tuning platform"

swingboy — Thu, 01 Oct 2026 17:06:14 +0000

It allows image input. Nice!

New comment by swingboy in "GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price"

swingboy — Tue, 29 Sep 2026 20:57:27 +0000

If you disable all other providers but DeepSeek in your OpenRouter guardrails, is that effectively the same thing?

New comment by swingboy in "Jeeves. Reasoning improves Jev-like decision models"

swingboy — Tue, 29 Sep 2026 14:31:47 +0000

Any good classifiers like this or Jev that support image input?

New comment by swingboy in "Show HN: Pac-Bench – How well can models one-shot a Pac-Man game?"

swingboy — Tue, 29 Sep 2026 09:30:15 +0000

Interesting the `high` effort on the Claude models. Did you find that going above that is unnecessary?

New comment by swingboy in "Sonnet 5.5"

swingboy — Mon, 28 Sep 2026 19:07:28 +0000

Is Opus still 2x usage of Sonnet after this? My Claude Code isn't showing that warning anymore when I look at /model.

New comment by swingboy in "DeepSeek Elastic Compute (DSec)"

swingboy — Sat, 26 Sep 2026 20:16:58 +0000

Is there a lab more innovative than DeepSeek? Imagine if they had the same compute resources that Anthropic and OpenAI have.

New comment by swingboy in "U.S. appeals court upholds designation of Anthropic as supply chain risk"

swingboy — Fri, 25 Sep 2026 17:50:39 +0000

I'm wondering how much was to push towards supporting Israel during the college campus protests and any future (at the time) conflict with Iran and it's proxies post October 7th like a lot of other Jewish million/billionaires such as Bill Ackman that typically supported Democrats or liberal causes. Brockman doesn't seem to have anything publicly stating he's Jewish but "Brockman" has English/German/Jewish origins and his mother's maiden name was Feldman.

New comment by swingboy in "Best LLM for every budget, updated daily"

swingboy — Thu, 24 Sep 2026 16:34:58 +0000

I guess it’s hard to actually calculate, but most people are probably on subscription plans with Anthropic or OpenAI and there’s not any good resources to estimate which models are the most “usage limit” efficient. I suppose you can sort of correlate raw API costs with how much usage a model would probably drain, but it would be nice if there was a reliable place to get that info.

New comment by swingboy in "People hooked on vapes try a new way to quit: cigarettes"

swingboy — Wed, 23 Sep 2026 17:02:53 +0000

Wellbutrin worked for me.

New comment by swingboy in "I built non-autoregressive decision models with RL a year ago"

swingboy — Sun, 20 Sep 2026 01:55:39 +0000

On a Mac it’s literally just option+dash.

New comment by swingboy in "We made Playwright 2x faster and 80% more token efficient"

swingboy — Sat, 19 Sep 2026 10:12:53 +0000

A lot of folks use Playwright for UI testing.

New comment by swingboy in "For AI leaders Doom is a form of hype"

swingboy — Mon, 14 Sep 2026 16:50:44 +0000

I’m apprehensive about using the recent hacking examples as proof that we’re “not even close to the wall” simply because such scenarios haven’t happened before. There’s a big difference between agents eventually hacking something because they just don’t get tired and can essentially brute force their way to a goal and super-intelligence. To be clear, I’m not saying the Hugging Face or Navier Stokes incidents aren’t impressive.

New comment by swingboy in "For AI leaders Doom is a form of hype"

swingboy — Mon, 14 Sep 2026 16:32:44 +0000

You don’t know the difference between train and test time? I asked for somebody from a lab.

New comment by swingboy in "For AI leaders Doom is a form of hype"

swingboy — Mon, 14 Sep 2026 16:29:45 +0000

Can anyone from a lab anonymously confirm they have or are insanely close to real-time weight modifications and still using the same architecture every other lab is using (transformers)?

New comment by swingboy in "AI recursive self-improvement might not come so quickly after all (August 2026)"

swingboy — Sun, 13 Sep 2026 19:47:42 +0000

There’s also the difference between a model recursively improving “itself” and improving itself via online learning.

The former being that these models are helping develop and train future models, but they might not veer too far off in architecture (yet).

The latter is a model being able to train/learn on the fly, in real time, permanently (not just in the current conversation/session), or in other words, adjusting/managing its own weights. But, it also seems like it would take an entire paradigm shift in model architecture from what most LLMs are built on, but I could be wrong.