Hacker News: ggcr

New comment by ggcr in "Muse Wonder"

ggcr — Mon, 12 Oct 2026 17:34:45 +0000

RL team hill-climbing this to 2nd AA place and this fast is crazy

How does it work on day-to-day workflows? does it translates well?

Language Designers Panel – Stroustrup, Guido van Rossum, Torgersen and Tracey

ggcr — Sat, 03 Oct 2026 20:25:34 +0000

Article URL: https://www.youtube.com/watch?v=HFcTjwfpHWQ

Comments URL: https://news.ycombinator.com/item?id=49947456

Points: 2

# Comments: 0

Grok TiddlyWiki, the definitive TiddlyWiki learning resource

ggcr — Mon, 28 Sep 2026 16:30:56 +0000

Article URL: https://groktiddlywiki.com/read/

Comments URL: https://news.ycombinator.com/item?id=49880591

Points: 1

# Comments: 0

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

ggcr — Thu, 24 Sep 2026 16:04:39 +0000

Oh yeah totally, good point. I remember there was controversy around Claude tokenizer generating more tokens lol

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

ggcr — Thu, 24 Sep 2026 15:56:12 +0000

What's impeding a lab from releasing its own new model, pricing it really low for the beautiful Pareto plot, accompanied by phrases VC love like "establishing a new frontier in cost", to just then raise prices back up?

New comment by ggcr in "GPT-6 Sol and Luna"

ggcr — Tue, 22 Sep 2026 18:30:34 +0000

Live notification in Codex:

> GPT-5.6-Sol is retiring. This conversation will automatically switch to GPT-6-Sol

I don't recall OAI retiring a model so early lol. Similar arch?

New comment by ggcr in "Introducing System One Models and Jev"

ggcr — Wed, 16 Sep 2026 16:25:35 +0000

Interesting. Perhaps I can see this being quickly adopted in LLMs-as-a-judge, where you normally need (a) a structured answer, say, with lots of different fields (metrics) and (b) you want the judge to be fast, not being a bottleneck.

New comment by ggcr in "I've operated petabyte-scale ClickHouse clusters for 5 years"

ggcr — Sat, 12 Sep 2026 13:50:45 +0000

The CTO himself reported 600 commits and 300 PRs each day. And they can do so because they have a massive CI [1]:

> Every day CI runs about 20..80 million tests in 600 commits and 300 pull requests

> Last year, ClickHouse spent 360 years of machine time for CI

I am no user of CH so I can't talk about their product. But we are talking about a company with 686 employees as per their LinkedIn, where ClickHouse is clearly the core of their business. Considering all of this, is 50+ commits a day that much?

[1] https://presentations.clickhouse.com/2026-openhouse-sf/great...

New comment by ggcr in "Apple Unveils iPhone Duo"

ggcr — Wed, 09 Sep 2026 18:49:11 +0000

> I am so used to using one hand to operate phones.

I even forgot about this. God damn it, I enjoyed the SE so much because of this.

New comment by ggcr in "iPhone Duo"

ggcr — Wed, 09 Sep 2026 18:13:32 +0000

Seems like a big engineering feat by Apple, but for real, who asked for this?

Edit: Is there that many demand for a foldable iPhone that justifies Apple making such engineering investment?

New comment by ggcr in "Navier-Stokes – Tristan Buckmaster [pdf]"

ggcr — Tue, 08 Sep 2026 09:48:47 +0000

Reminds me, kinda, to when Astra was launched and OpenAI announced an improvement to the bounded prime gap. Which BTW, Prof. Julia Stadlmann had published an independent result only a few days earlier

Stadlmann improved it from 246 to 240, OpenAI later claimed 186 I think?

Maybe someone can help clarify? I am no expert at all, but I can't help but see similarities.

[0] https://arxiv.org/abs/2608.31126

New comment by ggcr in "Navier-Stokes – Tristan Buckmaster [pdf]"

ggcr — Tue, 08 Sep 2026 07:26:53 +0000

> Let me make plain what I have said to colleagues in private: in view of this body of work, I believe Luis Martínez-Zoroa deserves a Fields Medal.

New comment by ggcr in "Show HN: GET Together – A social network where you don't need POST to Post"

ggcr — Mon, 07 Sep 2026 12:11:32 +0000

This would go hard if I was an OpenAI agent

Reading About the (2008) Financial Crisis: A 21-Book Review, by Andrew Lo (2012) [pdf]

ggcr — Sat, 05 Sep 2026 18:48:29 +0000

Article URL: https://www.stat.berkeley.edu/~aldous/157/Papers/lo.pdf

Comments URL: https://news.ycombinator.com/item?id=49579468

Points: 2

# Comments: 0

New comment by ggcr in "The efficient frontier of LLM inference"

ggcr — Wed, 02 Sep 2026 12:40:19 +0000

2026 has been the year where spec-dec has matured, it has been adopted by all big OSS engines and i'm sure it's present in quite a lot of inference providers as the default

I feel like P/D dissaggregation will be the next big one for providers, as prefill tends to be compute bound while decode mem bound which I guess each will have a different type of node

New comment by ggcr in "Nvidia agrees to acquire Hugging Face for $13B"

ggcr — Thu, 27 Aug 2026 06:29:16 +0000

Here is an employee of HF at a time answering how do they make money, supposedly [1]

> We make money via compute credits + Enterprise Hub + HF Pro subs [...]

I guess this plus custom inference deployments, external inference providers, partnerships with the big cloud AWS, Azure, etc.

[1] https://x.com/reach_vb/status/1928050126498713706

New comment by ggcr in ""Famous Deep Learning Papers", David Bau"

ggcr — Wed, 26 Aug 2026 17:54:26 +0000

Agree. I'm not into interpretability but I always try to follow what Bau and his team are up to.

"Famous Deep Learning Papers", David Bau

ggcr — Wed, 26 Aug 2026 12:44:26 +0000

Article URL: https://papers.baulab.info/

Comments URL: https://news.ycombinator.com/item?id=49448082

Points: 33

# Comments: 5

New comment by ggcr in "Ox-Alpha Is GLM?"

ggcr — Tue, 25 Aug 2026 07:31:11 +0000

Z.ai founders were one of the pioneers in MM-LLMs with Cog-VLM years ago, back when LLaVA emerged. I wouldn't be surprised if they added multi-modal capabilities

New comment by ggcr in "Ox-Alpha Is GLM?"

ggcr — Tue, 25 Aug 2026 06:36:38 +0000

Ziphu has that many resources to be able to serve capacity for 1 quadrillion tokens per day on Nous portal? My bet is that it's a Composer model from Cursor running on xAI cluster, they already did a Composer based on Kimi-K2.5