THE HACKER NEWS

kbyatnal
6 hours ago
Theranos.world
An interactive simulation of Elizabeth Holmes' desk. Open her MacBook, scroll her iPhone, run the Edison machine. Every text and email is real, from the court trial, parsed by Extend....
2 DAYS AGO
BOXSTREAM
Show HN: Rgpu – a PyTorch device whose tensors live on a remote GPU
A DAY AGO
SPECKX
The dawn of the age of the exoskeleton
6 HOURS AGO
LUKEED
Show HN: Pocketty – iPhone SSH terminal that pings you when an agent is blocked
2 DAYS AGO
JOEGIBBS
Show HN: TerrainSR – fast, realistic heightmap upscaling model
13 HOURS AGO
GEOX
Vanillin provides a sweet solution for chronic wound healing
12 HOURS AGO
SONABINU
Anne Carson wins Nobel Prize in literature 2026
4 DAYS AGO
SHAHIDHUSSAIN
15-year search for a band that charted once and vanished
7 HOURS AGO
MFIGUIERE
OpenAI annualised revenues $20B less than previously signalled
8 HOURS AGO
01-_-
4-hour battery storage is cheaper to install than gas turbines all across globe
2 HOURS AGO
DEATHG0D
Show HN: I built a fast spreadsheet engine for the web in Rust/WASM
md224
41 minutes ago
OpenAI, the Partition Principle, and Mathematics
By the time I got out of bed yesterday, I had three people ask me if I'd seen the OpenAI announcement that the Partition Principle does not (...)
image
ray__
3 hours ago
AI-ready biological data: $1.8B global commitment
Biohub, DOE, NIH, and partners will generate open, standardized data to train AI models that predict how cells respond to interventions....
bryanrasmussen
a day ago
A 5.3M-year-old deep-sea whale necropolis in the Diamantina Zone
Researchers uncovered an enormous deep-sea accumulation of whale remains in the southeastern Indian Ocean, showing long-term, specialized ecosystems and an extensive fossil record that offer...
AnneWodell
8 hours ago
Step 5 Preview, a 1M-context MoE from StepFun, shows up on OpenRouter
Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). $1 per million input tokens, $2.70 per mi...
soheilpro
2 days ago
I hired an illustrator to draw my house. Now it's my Home Assistant dashboard
A flickering fireplace, a softly lit garden, and a smart home the whole family finally enjoys....
bookofjoe
3 hours ago
ADHD as a circadian rhythm disorder: evidence and implications for chronotherapy (2025)
Accumulating evidence indicates that circadian rhythm dysfunction is a clinically significant and highly prevalent phenotype in a substantial subgroup of ind......
speckx
11 hours ago
Beauty in DVD Menus
The best part of digital video disc isn't the video....
ck2
a day ago
Man discovers his parents' coffee machine used 1TB of data in 10 days
The man said they discovered that their parents’ coffee maker had generated far more traffic than they expected....
https://www.dexerto.com/entertainment/man-discovers-his-parents-coffee-machine-used-1tb-of-data-in-10-days-3416399/
saikatsg
a day ago
DuckDB Ducklake
DuckLake is an integrated data lake and catalog format - duckdb/ducklake...
stared
12 hours ago
I gave Opus 5.5 one prompt and six hours to visualize Invisible Cities
Claude Opus 5.5 vs GPT-6 Astra one-shot Invisible Cities by Italo Calvino in Three.js: AI design, UX, data visualization and explorable explanations....
andreynering
10 hours ago
Orkut.com
You deserve better from social media. You deserve a more human internet. Help Orkut to bring back what really matters: people, communities, and genuine connection....
benbreen
2 days ago
The Slow Formation of Durable Software
How a group of historians took years to imagine Zotero, the research application that would eventually be used by millions...
sashank_1509
17 hours ago
OpenAI withdraws three mathematical results
We've updated our GitHub math repo with 6 new Lean formalizations, 19 modifications, and 3 withdrawals. The repo now has ~42% top-line results formalized. We will continue to update the repo...
higeorge13
5 hours ago
We have LLMs now. Why are the docs still wrong?
Keeping documentation in sync with code takes more than generating text. My take on LLMs, stale docs, and what I am building with Amendary....
danny00
11 hours ago
New CRAM method offers giant boost to compressed memory reads
The new method allows access using standard memory semantics instead of as a block device, drastically improving performance....
brainless
11 hours ago
Sub-1-Bit LLM Compression via Latent Factorization
Official implementation of LittleBit (NeurIPS 2025) and its follow-up LittleBit-2 (ICML 2026) - SamsungLabs/LittleBit...