LLM stories
Large language models have moved from research curiosity to production infrastructure. This topic focuses on LLM APIs, fine-tuning, RAG pipelines, agent frameworks, and the economics of running models at scale.
HN discussions here tend to be unusually practical — benchmark comparisons, cost breakdowns, and real deployment war stories from teams shipping LLM features.
889 stories archived · Page 4 of 30
RSS feed for LLMThe American Religion of Self-Storage Facilities
newyorker.com · 51 points · 455 comments
A Letter from a Machine Learning Engineer
nemin.hu · 3 points · 0 comments
Facing Outrage, Flock Got Help from Group That Uses AI to Rally Support
theintercept.com · 3 points · 0 comments
GLM Built Its Own Inference Infrastructure
z.ai · 192 points · 285 comments
OpenAI models secretly generate instructions to ignore constraints
alignment.openai.com · 42 points · 37 comments
Breaking the 1.58-bit Barrier for Ternary LLMs
arxiv.org · 241 points · 41 comments
Barndoor acquires Diaphora, creators of open-source workflow runtime Frags
barndoor.ai · 7 points · 3 comments
You can run Git on object storage if you re-make packfiles
tigrisdata.com · 31 points · 35 comments
Fedora 45 beta drags the Linux console into the 21st century
theregister.com · 13 points · 3 comments
PS5 Linux lead quits: "a bunch of noobs using LLMs" that "they don't understand"
frvr.com · 313 points · 225 comments
OmnisBench, a re-gradable, open LLM routing benchmark on fresh tasks
github.com · 3 points · 0 comments
Learning Programming in an Age of LLMs
blog.ploeh.dk · 83 points · 195 comments
Deep Seek v4.1 M5 Max at 17 tokens/s
github.com · 13 points · 2 comments
Musk proposes adversarsial peer review for AI Safety
twitter.com · 4 points · 2 comments
Slow OpenAI Inference on AWS Bedrock
2 points · 2 comments
Micron Shows Off 512GB DDR5 Rdimm: 12TB per Dual-Socket Server at 9,200 MT/S
storagereview.com · 4 points · 0 comments
Iran war has cost on average $246M per day in its first five months
cnbc.com · 25 points · 9 comments
Learning to solve hard problems in RL for LLMs by never giving up
mnoukhov.github.io · 119 points · 9 comments
Why I'm still bearish on LLMs after Navier-Stokes
dank.systems · 495 points · 652 comments
AirmailAI, a BYOK LLM chat app with a browser extension backend
airmailai.net · 3 points · 7 comments
A search-and-inference database from scratch in pure Zig
antfly.io · 30 points · 19 comments
Building your first LLM API call in Python (step by step)
heymeraki.substack.com · 8 points · 2 comments
LoongForge-Train LLMs, VLMs, diffusion and embodied models, faster
github.com · 2 points · 0 comments
The CSS Zen Garden dream, finally shipped
josprague.com · 81 points · 113 comments
GRP-Obliteration: Unaligning LLMs with a Single Unlabeled Prompt
arxiv.org · 15 points · 9 comments
MSI XpertStation WS300 Thermals: Why a 1,300W GB300 Doesn't Throttle on a Desk
storagereview.com · 4 points · 0 comments
Let agent read files with secrets while redacting values for LLM contex
github.com · 3 points · 1 comments
The Inference Hardware Revolution of 2026
spectrum.ieee.org · 140 points · 22 comments
KaozKit – JavaScript LLM agents on an engine built for microcontrollers
github.com · 3 points · 3 comments
Sovereign: A Unified GPU Inference Substrate (Fractal Memory, Manifold Routing)
github.com · 5 points · 2 comments