Claude Sonnet 4 now supports 1M tokens of context(anthropic.com)
1264 points by adocomplete | 668 comments
Anthropic product announcement that Claude Sonnet 4 now supports a 1 million token context window.
Hacker News on
Stories posted this day, by points. Rank history starts with the October 2026 rebuild.
Show HN: Building a web search engine from scratch with 3B neural embeddings(blog.wilsonl.in)
578 points by wilsonzlin | 97 comments
Wilson Lin's blog post describing how he built a web search engine from scratch in two months using 3 billion neural embeddings.
Search all text in New York City(alltext.nyc)
520 points by Kortaggio | 101 comments
A search tool that lets users find text appearing on New York City streets, extracted via OCR from street-view imagery.
What's the strongest AI model you can train on a laptop in five minutes?(seangoedecke.com)
476 points by ingve | 173 comments
Sean Goedecke's blog post on training the best small GPT-style model possible on a MacBook Pro in five minutes, landing on a ~1.8M-parameter transformer trained on TinyStories.
Why are there so many rationalist cults?(asteriskmag.com)
459 points by glenstein | 695 comments
Asterisk Magazine essay by Ozy Brennan examining why the Rationalist community tends to spawn cult-like high-demand groups.
Monero appears to be in the midst of a successful 51% attack(twitter.com)
458 points by treyd | 209 comments
An X post by Charles Guillemet claiming Monero is undergoing a successful 51% attack by the Qubic mining pool, which has gained majority hashrate and caused a chain reorganization.
StarDict sends X11 clipboard to remote servers(lwn.net)
441 points by pabs3 | 290 comments
LWN article reporting that the StarDict dictionary app on Debian sends X11 text selections over unencrypted HTTP to remote Chinese dictionary servers.
![Screenshot of arxiv.org: GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models [pdf]](https://hn.urlbox.com/shots/renders/2025/8/12/68eb07e3-6f81-44a7-9646-4891d3adb444.jpg)
GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models [pdf](arxiv.org)
382 points by SerCe | 71 comments
arXiv technical report from Zhipu AI and Tsinghua introducing GLM-4.5, a 355B-parameter open-source MoE language model for agentic, reasoning, and coding tasks.
Australian court finds Apple, Google guilty of being anticompetitive(ghacks.net)
335 points by warrenm | 126 comments
A ghacks.net news article reporting that an Australian court found Apple and Google engaged in anticompetitive conduct.
Training language models to be warm and empathetic makes them less reliable(arxiv.org)
334 points by Cynddl | 333 comments
arXiv paper finding that fine-tuning language models to be warm and empathetic raises error rates and makes them more sycophantic on safety-critical tasks.
GitHub is (again) having issues(githubstatus.com)
321 points by rock_artist | 195 comments
GitHub status page incident report about increased failure rates affecting search on GitHub.
Ashet Home Computer(ashet.computer)
294 points by todsacerdoti | 85 comments
Go 1.25 Release Notes(go.dev)
194 points by bitbasher | 50 comments
![Screenshot of media.ccc.de: WHY2025: How to become your own ISP [video]](https://hn.urlbox.com/shots/renders/2025/8/12/33a4adb2-ddde-44d1-bb5e-7567a64e77e4.jpg)
Journaling using Nix, Vim and coreutils(tangled.sh)
186 points by icy | 54 comments
500 days of math(gmays.com)
146 points by gmays | 80 comments
Weathering Software Winter (2022)(100r.co)
113 points by todsacerdoti | 47 comments
Radicle 1.3.0(radicle.xyz)
68 points by Skinney | 5 comments
![Screenshot of youtube.com: Architecting large software projects [video]](https://hn.urlbox.com/shots/renders/2025/8/14/4ec3055f-fee0-4baf-905e-ba60f04192b3.jpg)
Print, a one-line BASIC program(10print.org)
61 points by NKosmatos | 17 comments
From Here?(dirtyfeed.org)
49 points by hooboy | 6 comments
Scapegoating the Algorithm(asteriskmag.com)
34 points by fmblwntr | 17 comments