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.
891 stories archived · Page 12 of 30
RSS feed for LLMMoonshot in talks with Microsoft, Amazon, Google over K3 inference revenue share
reuters.com · 6 points · 1 comments
RAG Is Simpler Than You Think
lighthousenewsletter.com · 153 points · 218 comments
Revolut Joins Global Stablecoin Race with Euro-Backed Token
bloomberg.com · 9 points · 1 comments
France, Saudi Arabia agree on €6B Dragon Ball Z theme park project near Paris
reuters.com · 10 points · 1 comments
A more nuanced view of LLMs
anarc.at · 6 points · 0 comments
LLM-powered webapp to build LLM-powered webapps
codeplusequalsai.com · 2 points · 2 comments
What is one simple thing LLMs are insanely bad at?
25 points · 91 comments
Darkbloom (AI inference on idle Macs) – security audit with PRs submitted
gist.github.com · 3 points · 0 comments
Dependencies in the LLM API Reseller Ecosystem
arxiv.org · 4 points · 0 comments
Training LLMs to write tools generalized beyond self use
arxiv.org · 6 points · 0 comments
A batch settlement layer for tokenized NYSE stocks
github.com · 3 points · 0 comments
Beyond China's humanoid robots, a quieter machine revolution is unfolding
bbc.com · 7 points · 0 comments
Slash-tokens – know LLM cost before the call leaves your machine
github.com · 4 points · 5 comments
Microsoft Copilot Cowork Controlled by Attacker, Bypassing Sandbox
promptarmor.com · 3 points · 0 comments
Cross-vendor byte-identical inference for a 72B LLM (AMD MI300X vs. Nvidia H100)
zenodo.org · 8 points · 0 comments
I missed the moving blocks, so I built a real Linux disk defragmenter
github.com · 3 points · 69 comments
Dribbling the AI Watermark Directly In-Prompt
explore-exploit.com · 2 points · 0 comments
Red-team LLM reasoning and agent actions (honest scoring, local-first)
github.com · 4 points · 0 comments
Turn any website into a CLI for AI agents (142x fewer tokens than HTML)
github.com · 3 points · 0 comments
Open-source AMDGCN kernels for optimizing LLM inference
github.com · 5 points · 0 comments
My experience with LLM-assisted tools in software development
ounapuu.ee · 3 points · 0 comments
Japan enlists 1,800 people to drag 360-tonne castle keep using ropes and rollers
theguardian.com · 47 points · 3 comments
Screen memory without screenshots, just text to Markdown
github.com · 61 points · 27 comments
Thomson Reuters Launches Its Own Frontier Model
thomsonreuters.com · 87 points · 54 comments
SK hynix runs out of replacement SSDs and defaults to purchase price refunds
tomshardware.com · 4 points · 2 comments
Bookshelf – Self-hosted eBook library that runs on object storage
github.com · 172 points · 64 comments
LLMs could control their host machines by exploiting inference engines
boydkane.com · 27 points · 108 comments
I built a lite LPU that can do inference on Karpathy's MicroGPT
lpulite.com · 18 points · 3 comments
Public services are increasingly strained by LLM-written appeals for benefits
arxiv.org · 41 points · 87 comments
A Server Lost Power at 00:32. We Found Out at 08:18
danubedata.ro · 10 points · 9 comments