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 11 of 30
RSS feed for LLMDoctors are finally learning to manage antidepressant withdrawal
newscientist.com · 193 points · 369 comments
AI Engineer Notebooks – free, framework-free RAG/agents/evals on Colab
github.com · 111 points · 15 comments
Storm Summoner, a MIDI controller for effects pedals
kabaragoya.com · 23 points · 8 comments
Benchmarking Pocket-Scale Inference
artificialanalysis.ai · 82 points · 18 comments
Fewer Americans Pay to Use LLMs Than Still Pay to Play World of Warcraft
wjamesau.substack.com · 27 points · 17 comments
Gemma 4 E2B inference in 700 lines of C
github.com · 9 points · 0 comments
Dad's Custom Atari Peripherals – By Jim Trageser
goto10retro.com · 8 points · 21 comments
Backprompter – create, test, and deploy agents without a back end
backprompter.com · 3 points · 0 comments
Nanointerpret – LLM Interpretability Playground
nanointerpret.pages.dev · 5 points · 1 comments
Millnew AI – Explaining time-series models using local LLMs and Captum
robinhood-demo.streamlit.app · 5 points · 1 comments
Drag-and-drop seating charts for weddings and events
planseats.com · 5 points · 0 comments
Changes to Sourcehut's terms of service regarding LLMs
sourcehut.org · 129 points · 43 comments
Meta's self-inflicted resignation-wave
blog.pragmaticengineer.com · 27 points · 1 comments
Debian weighs eight options in vote on LLM usage
lwn.net · 5 points · 0 comments
Use GLM-5.3 in Cursor today via tokengo API
tokengo.com · 3 points · 2 comments
Ultrafast Frontier Inference – Cerebras Hot Chips 2026
cerebras.ai · 9 points · 0 comments
Railo – Deterministic security patch bot using AST and Z3 (no LLMs)
railo.dev · 4 points · 1 comments
Timber (iOS) and Timber Tabs (Mac) – best offline read-aloud LLMs
timberreader.com · 12 points · 1 comments
I get 25 deeply researched ideas from 19 agents with one single prompt
github.com · 2 points · 0 comments
DuckDB speed on MySQL without a new storage engine
percona.community · 3 points · 0 comments
Llmcanvas.chat Tree-based LLM chat on an infinite canvas
llmcanvas.chat · 4 points · 0 comments
Launch HN: Risklytics (YC S26) – Insurance brokerage for frontier tech companies
risklytics.ai · 24 points · 25 comments
Prompt Builder – Free Templates
promptbuilder.space · 20 points · 0 comments
France reaches 94.9% fiber coverage in 2026
cartefibre.arcep.fr · 308 points · 239 comments
An 8.6 GB model that serves only 7 requests a second
mapathak-commits.github.io · 10 points · 4 comments
ModelMRI – see inside a local LLM, VLM or robot policy while it runs
github.com · 4 points · 0 comments
PyCon26: LLM Governance, Guardrails, and Presidio When the Guardrail Leaks PII
petrostechchronicles.com · 6 points · 1 comments
PepsiCo Scraps Coverage of Weight-Loss Drugs for Employees
bloomberg.com · 7 points · 0 comments
Axera AX8850 LLM running ggufs
github.com · 5 points · 0 comments
Code Stitcher – Apply any LLM output to your local codebase
3 points · 0 comments