H3-metal – Native MiniMax-H3 inference for Apple Silicon
swyx
Show HN: Mcptoon – MCP CLI client that cuts tool discovery tokens by 97%
mcptokensaver
Chicken Scheme 6.0
eatonphil
LFM2.5 2.6B model competitive with 4x larger models
nateb2022
Show HN: Scroll through all 43252003274489856000 Rubik's Cube states
Alen123
Show HN: Needle2: 14MB agentic LLM for phones, wearables, smart home and robots
HenryNdubuaku
The “mechanical miracle” that ruined Mark Twain’s life
benbreen
Recycle – Floppydisks
calvinmorrison
Mark Zuckerberg attacks 'closed' AI rivals as Meta returns to open models
root-parent
Hyperspace
swyx
Stowaway – Take the window seat on any plane or satellite overhead
thunderbong
Sonic Pi v5
samaaron
Rust SIMD on the GPU
sagacity
World Train Map – 1247 train routes around the world
Flightmussy
DeepSeek: Reverse Engineering an AI Assistant by Interviewing Itself
ms7892
What's the best programming language for coding agents?
chaychoong
Publishing Schematics Before “Open Source” Was a Word
extralongdivisi
Confessions of a Long-Distance Sailor
AntiRush
Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows
riordan
Squeak 6.1
fniephaus
Choral: Choreographic Programming for Java
dplyukhin
Tail-call optimization in C is relatively recent (2025)
prakashqwerty
Humanising LLM Outputs Is Dumb
kuberwastaken
Exploiting System Management Mode with a very long interrupt
WhiteDawn
Once More in Triple Time
Petiver
Parametron: 50s Japanese computer that uses neither transistors nor vacuum tubes
xeonmc
Launch HN: Stoa Markets (YC S26) – A Marketplace for GPUs and AI Servers
erenberke
Updated GPG Key for Signing Firefox and Thunderbird Releases
csmantle
Exploring Claude/GPT Knowledge Cutoffs and Pre-Training Timelines
sshh12
Half of Europe's towns and villages have fewer residents than 60 years ago
Pamar
49253250