newest
Open in hackernews
Physicists observe a new form of magnetism
https://news.mit.edu/2025/physicists-observe-new-form-magnetism-0605
3
•
chmaynard
•
1d ago
What the best AV1 encoder in 2025?
https://catskull.net/libaom-vs-svtav1-vs-rav1e-2025.html
1
•
andsoitis
•
45s ago
•
0 comments
TinyWM – The tiniest (X11) window manager
https://github.com/mackstann/tinywm
1
•
peter_d_sherman
•
2m ago
•
0 comments
Show HN: Instant video edits with local Whisper models (macOS)
https://cutword.com
2
•
jelled
•
3m ago
•
0 comments
Show HN: Sip Bridge for Galene
https://github.com/jech/galene-sip
1
•
jech
•
5m ago
•
0 comments
Teenage Engineering lets you pick the price for their OP-1f
https://teenage.engineering/
2
•
Fraterkes
•
7m ago
•
0 comments
LongCodeBench: Evaluating Coding LLMs at 1M Context Windows
https://arxiv.org/abs/2505.07897
1
•
PaulHoule
•
9m ago
•
0 comments
Hypermultiplexed integrated photonics–based optical tensor processor
https://www.science.org/doi/10.1126/sciadv.adu0228
1
•
hbartab
•
10m ago
•
0 comments
Ask HN: Has flawless grammar become a liability in the age of AI?
1
•
amichail
•
10m ago
•
2 comments
The Rule of Idiots
https://chrishedges.substack.com/p/the-rule-of-idiots
2
•
chmaynard
•
11m ago
•
0 comments
You could automate (some) boilerplate Go error handling with a formatter
https://utcc.utoronto.ca/~cks/space/blog/programming/GoErrorBoilerplateViaFormatter
1
•
ingve
•
12m ago
•
0 comments
Using Kodus to connect open source LLMs to your code reviews
https://docs.kodus.io/cookbook/en/novita
1
•
edvaldodfreitas
•
14m ago
•
0 comments
LLMs in Public Health
https://joshuaharrissite.substack.com/p/llms-in-public-health-part-1
1
•
jah242
•
14m ago
•
0 comments
I published this blog post 3 years ago, still no answer to the entropy question
https://quickchat.ai/post/tokens-entropy-question
1
•
piotrgrudzien
•
14m ago
•
1 comments
2014 Eugene Jarvis describes Robotron 2084 in detail [video]
https://www.youtube.com/watch?v=8AbiBOK8jZ0
1
•
mosura
•
15m ago
•
0 comments
LLM-Powered Industrial Sabotage
https://ianreppel.org/llm-powered-industrial-sabotage/
1
•
hbartab
•
15m ago
•
0 comments
Deezer reveals 18% of all new music uploaded to streaming is fully AI-generated
https://newsroom-deezer.com/2025/04/deezer-reveals-18-of-all-new-music-uploaded-to-streaming-is-fully-ai-generated/
1
•
bookofjoe
•
15m ago
•
1 comments
Show HN: MVPster.com - Monetize abandoned projects
https://mvpster.com
1
•
mhnnnn
•
15m ago
•
0 comments
We Made Frontier AI Models Compete in a Game of Diplomacy. Here's Who Won
https://every.to/p/diplomacy
1
•
babushkaboi
•
18m ago
•
0 comments
AI is a gamechanger for TLA+ users
https://buttondown.com/hillelwayne/archive/ai-is-a-gamechanger-for-tla-users/
1
•
mpweiher
•
18m ago
•
0 comments
Unrestricted Browser Networking: Raw TCP Sockets, Modern TLS, and CORS-Free HTTP
https://developer.puter.com/blog/unrestricted-browser-networking-raw-tcp-sockets-modern-tls-and-cors-free-http/
1
•
thunderbong
•
18m ago
•
0 comments
How US and Israel Could Destroy Iran's Nuke Program
https://www.bloomberg.com/opinion/articles/2025-06-06/us-israel-could-destroy-iran-s-nuclear-sites-if-it-rejects-trump-deal
1
•
Bostonian
•
19m ago
•
1 comments
Vibe Debt: The Hidden Crisis Coming for Your Product Team
https://medium.com/@dan.mark.cohen/vibe-debt-the-hidden-crisis-coming-for-your-product-team-262dd3fa4ad4
2
•
matthewsinclair
•
20m ago
•
0 comments
Context is all you need
https://fakepixels.substack.com/p/context-is-all-you-need
1
•
MukundMohanK
•
22m ago
•
0 comments
GZDoom: OpenGL and Vulkan source port based on the DOOM engine
https://github.com/ZDoom/gzdoom
1
•
klaussilveira
•
22m ago
•
0 comments
Mental Refactoring for the AI Age
https://zach.codes/p/mental-refactoring-for-the-ai-age
1
•
zackify
•
22m ago
•
0 comments
Process improvement is more important than metrics
https://khanin.info/en/blog/244
1
•
daniilkhanin
•
24m ago
•
1 comments
Show HN: Kefir – Native macOS menubar app for KEF wireless speakers
https://github.com/melonamin/Kefir
1
•
melonamin
•
27m ago
•
0 comments
It turns out you can train AI models without copyrighted material
https://www.engadget.com/ai/it-turns-out-you-can-train-ai-models-without-copyrighted-material-174016619.html
1
•
speckx
•
29m ago
•
0 comments
Symbase – A Symbolic Logic OS You Can Run in the Browser
https://fr3action.com
2
•
memv
•
30m ago
•
1 comments
Mono-JSX: JSX runtime for server side
https://github.com/ije/mono-jsx
1
•
ambigious7777
•
32m ago
•
0 comments