ASRock BC-250: Building the Budget Steam Machine
Article URL: https://plug-world.com/posts/2026/asrock-bc250-the-budget-steam-machine/ Comments URL: https://news.ycombinator.com/item?id=49140213 Points: 15 # Comments: 1
Communities 6915 stories
The Prospects for 128 Bit Processors ( John Mashey SGI 1995)
Article URL: https://yarchive.net/comp/128bit.html Comments URL: https://news.ycombinator.com/item?id=49140174 Points: 8 # Comments: 0
A Chess Training App for Imitation Learning
LLMs learn faster if we first pretrain them to imitate dense teacher-forced examples. I speculated that this would work on humans too, so I built a…
MUD as AI Evaluation and LLM-judge distortion in ways aggregate κ misses
A group of friends and I spent the last several months running an experiment in our free time to determine if a MUD would be a…
Expert-only IQ3 requant of DeepSeek-V4-Flash-0731: better KLD than UD-IQ3_S, 1.4x decode on a CPU-spill rig
Hey all, tldr / who this helps: you run a mixed multi-GPU box where the experts spill to RAM, and you want to stay in the…
Persistent State Machines: LLM Attention with INT4 In-Memory Cells
Article URL: https://zenodo.org/records/21753002 Comments URL: https://news.ycombinator.com/item?id=49140080 Points: 3 # Comments: 0
Constitutional Midtraining: Content Presence Drives Alignment Gains
A more accessible, much shorter version of our paper that goes by the above title. Paper here. Code and benchmarks here. Data and models here. Would…
Walsh: Multi-agent research pipeline with risk manager that can veto trades
Article URL: https://github.com/ats4321/walsh Comments URL: https://news.ycombinator.com/item?id=49139865 Points: 3 # Comments: 0
Why are almost all new benchmarks and leaderboards coding focused?
I know in in this community LLM's are generally used for coding but there are other usecases besides coding and those usecases should be tested too.…
The Greenhouse and the Lens: Two Modes of Agentic AI Work
Article URL: https://www.brethorsting.com/blog/2026/08/the-greenhouse-and-the-lens-two-modes-of-agentic-ai-work/ Comments URL: https://news.ycombinator.com/item?id=49139826 Points: 3 # Comments: 0
How well do multiple GPUs scale for LLM inference? (Trying to understand the basics)
Hi everyone, I’m fairly new to the multi-GPU side of local LLMs and I’m trying to understand how inference actually scales across multiple GPUs. Suppose I…
Kobuk the Destroyer – Tales from the Wild, Unseen World of Test Engineering
Article URL: https://wwnorton.com/books/9781324051299 Comments URL: https://news.ycombinator.com/item?id=49139796 Points: 3 # Comments: 1
Why Is Europe Burning?
Article URL: https://www.newyorker.com/news/the-lede/why-is-europe-burning Comments URL: https://news.ycombinator.com/item?id=49139790 Points: 13 # Comments: 1
We must go deeper – Inception style experience with local AI
https://preview.redd.it/6aodizfzpugh1.png?width=1252&format=png&auto=webp&s=5749d622a050de8c325093549bb0cd31b1050cd9 I am trying to learn Japanese so I vibe coded scripts that convert book page images into a website with Kokoro TTS voiceovers and contextual…
A Uiua Type System
Article URL: https://www.uiua.org/blog/a-uiua-type-system Comments URL: https://news.ycombinator.com/item?id=49139741 Points: 3 # Comments: 1
RFC 10015: Deprecating Obsolete Key Exchange Methods in TLS 1.2 and DTLS 1.2
Article URL: https://www.rfc-editor.org/rfc/rfc10015.html Comments URL: https://news.ycombinator.com/item?id=49139711 Points: 4 # Comments: 0
Moore's Law for Everything (2021)
Article URL: https://moores.samaltman.com/ Comments URL: https://news.ycombinator.com/item?id=49139699 Points: 4 # Comments: 0
For Local, what are your minimum good or usable tokens per second, for both promp processing and text generation?
Hello guys, hoping you're doing fine. Lately with all the new models, and how popular is offloading, what are your min good or usable t/s for…
A stray commit buried multiple levels deep cost me months
Article URL: https://www.droppedasbaby.com/posts/db-commits/ Comments URL: https://news.ycombinator.com/item?id=49139562 Points: 23 # Comments: 7
RLVR that rewards red teaming the training environment
Epistemic status: seeing what sticksI've been thinking pretty obsessively about how to make sure the Hugging Face incident doesn't happen again. I don't work at a…
Unraveling the mysteries of habit formation
Article URL: https://www.kyoto-u.ac.jp/en/research-news/2026-07-28 Comments URL: https://news.ycombinator.com/item?id=49139383 Points: 11 # Comments: 5
Morph (YC S23) Is Hiring Member of Technical Stuff
Article URL: https://www.ycombinator.com/companies/morph/jobs/0Z8vI3K-member-of-technical-staff Comments URL: https://news.ycombinator.com/item?id=49139352 Points: 0 # Comments: 0
"Everyone Has Been Sold a Lie" on AI
Article URL: https://www.youtube.com/watch?v=pHcZpvIfho0 Comments URL: https://news.ycombinator.com/item?id=49139325 Points: 9 # Comments: 0
Anime User Interfaces
Article URL: https://animeuserinterface.tumblr.com Comments URL: https://news.ycombinator.com/item?id=49139304 Points: 26 # Comments: 3