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.
898 stories archived · Page 21 of 30
RSS feed for LLMLaunch HN: EdotEnv (YC S26) – Quant Trading RL Envs to Teach LLMs Research
edotenv.com · 24 points · 34 comments
Microsoft Tells Engineers 'Tokenmaxxing Is Not What We Are Optimizing For
404media.co · 11 points · 0 comments
MariaDB: Promote getting to 10k GitHub stars in server log and client prompt
github.com · 42 points · 23 comments
Bundling Bioweapons with Vite
github.com · 10 points · 1 comments
BubbleHub - a local runtime and hosting for LLM agents
github.com · 2 points · 0 comments
Clai – AI for the command line (stdin → LLM → stdout)
github.com · 3 points · 0 comments
Moorage – a better way to mount and view MTP devices in macOS
github.com · 3 points · 0 comments
XTokenChecker – Verifies model identities of your AI gateway
xtokenchecker.com · 2 points · 0 comments
Simple self-hosted LLM assistant with user-steered compounding context
github.com · 3 points · 0 comments
The LLMs Problems
2 points · 1 comments
AI-Generated Images Discourage Me from Reading Your Blog
nelson.cloud · 756 points · 465 comments
Why Large Language Models Fail at Tabular Prediction
arxiv.org · 107 points · 33 comments
Homebench – Benchmark local LLMs for speed, memory, and quality
github.com · 59 points · 8 comments
Does AI research need "world models" more than bigger LLMs?
2 points · 2 comments
LLMs Can't Jump
openreview.net · 18 points · 3 comments
Gmail support for sending from third-party email addresses ends January 2027
support.google.com · 20 points · 8 comments
Windows XP 2002 for the Itanium: Unbridled rage
virtuallyfun.com · 129 points · 85 comments
LLMs reward expertise
seangoedecke.com · 679 points · 573 comments
Company Offering Printed Books to Train AI Stops After 404 Media Coverage
404media.co · 8 points · 0 comments
11-Node Agentic RAG with MCP and PII Shield Under 512MB RAM
agentic-rag-financial-parser.onrender.com · 5 points · 0 comments
A Chinese LLM attacked our lab, so we made it work for us
jesta.ai · 12 points · 6 comments
TokenMaxxer – track every AI token you spend across your coding tools
tokenmaxxer.xyz · 5 points · 0 comments
Critical CVE issued for hallucinated SQLite vulnerability
research.jfrog.com · 146 points · 372 comments
AirLLM 70B inference with single 4GB GPU
github.com · 63 points · 84 comments
Nightcrawler – A local AI pentesting agent running on a smartphone
github.com · 113 points · 35 comments
AQ – a from-scratch 1B academic LLM by a 2-person team in India
huggingface.co · 2 points · 0 comments
Prevent cognitive debt by manually retyping LLM-generated code
ankursethi.com · 131 points · 443 comments
Company Offering Printed Books to Train AI Stops After 404 Media Coverage
404media.co · 3 points · 0 comments
Do Codex skills save tokens? A six-run task-size benchmark
codex-howto-benchmark.nguyenvantamdk2.chatgpt.site · 4 points · 0 comments
Have LLMs Plateaued?
6 points · 29 comments