All topics

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.

895 stories archived · Page 15 of 30

RSS feed for LLM

An approach to vulnerability observability of Go services leveraging Prometheus

github.com · 3 points · 0 comments

Zuckerberg encouraged growth over child safety, ex-Meta executive testifies

reuters.com · 87 points · 22 comments

Claude Code adds new "concise" output style setting

twitter.com · 9 points · 0 comments

Turning websites into micro CLIs for Claude Code to save on tokens

github.com · 4 points · 2 comments

Pressed Penny Machine Map

pennypresses.net · 70 points · 53 comments

Servercoin- The first decentralized cloud marketplace with reviews

2 points · 0 comments

CT RailTime, a better way to travel on CTrail

ctrailtime.com · 2 points · 0 comments

Has anyone shipped a self-modifying application with LLMs?

4 points · 13 comments

LLM-as-a-Verifier Plugin for DeepSeek Harness

github.com · 2 points · 0 comments

X's algorithm feeds off ragebait and impacts Democrats more, study finds

404media.co · 19 points · 4 comments

HRAG – Hybrid RAG on €116/month of Hetzner, officially benchmarked

hrag.app · 7 points · 0 comments

Frugal Tokens – explore costs and usage across coding agents

demo.frugaltokens.com · 20 points · 10 comments

Extensible Software in the age of LLMs

jeremymorrell.dev · 64 points · 88 comments

LLM-Shield-Proxy Zero-Egress PII Streaming Proxy (55MB RAM)

github.com · 4 points · 1 comments

Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces

arxiv.org · 25 points · 9 comments

BSDun – FreeBSD ELF support for the Linux kernel

gitlab.com · 6 points · 2 comments

The Economics and Engineering of On-Premises LLMs

cacm.acm.org · 6 points · 1 comments

LLMs .what do you smoke beforehand?

5 points · 7 comments

The most influential economist is oddly unconvincing

economist.com · 13 points · 2 comments

200B Tokens Later: A Month of Letting AI Agents Decompile MW2

momo5502.com · 7 points · 3 comments

Deft Writing, an AI lab for non-slop LLMs

deftwriting.com · 4 points · 0 comments

Tracelint – a linter for AI agent traces, no LLM judge

github.com · 3 points · 0 comments

I benchmarked LLMs on predicting knife steel properties

github.com · 2 points · 0 comments

CDC: U.S. kindergarten vaccine exemptions hit record high as coverage falls

abc45.com · 9 points · 0 comments

ChatGPT has almost stopped citing Reddit

promptwatch.com · 32 points · 19 comments

Leviath – Structured context regions for LLM agents, in one Rust binary

leviath.dev · 2 points · 0 comments

Micron, SK Commit Billions to RAM Capacity, but Almost Nothing Lands Before 2028

storagereview.com · 21 points · 3 comments

Declarative-forms – await an object the way prompt() awaits a string

wolfoo2931.github.io · 6 points · 2 comments

If you're burning tokens at scale, what are you using them for?

4 points · 3 comments

LLM City – 3D render of all Kimi K3's weights as 2.5mm tiles

magik.net · 22 points · 7 comments