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.

898 stories archived · Page 23 of 30

RSS feed for LLM

Why we write our own C and C++ inference engines

localai.io · 51 points · 51 comments

A fundamental flaw leaves LLMs strikingly vulnerable to attack

technologyreview.com · 8 points · 1 comments

Prompt-scrub – local-first PII redaction for LLM prompts and responses

nanocollective.org · 4 points · 2 comments

13 Models and 4 Agents on SWE Tasks: Go, Java, Python, Rust, TS

swe-rebench.com · 27 points · 15 comments

BitBang – Reach machines behind NAT from a browser, no account

github.com · 39 points · 39 comments

Active inference world models: From finding a mug to making tea

cpnslab.com · 3 points · 0 comments

What are you using for LLM inference in production?

3 points · 0 comments

Yann LeCun's $1B Bet Against LLMs [Part 1] [video]

youtube.com · 16 points · 3 comments

Claude Opus 5 jailbreak with a 3-word prompt

twitter.com · 23 points · 4 comments

Exploring the "Dario and Amanda" Prompt

alec.is · 75 points · 17 comments

Can LLMs have an "accent" like second language speakers?

3 points · 0 comments

ChatGPT Optimizes Its Agent Loop: Harness, API, and Inference

blog.bytebytego.com · 4 points · 0 comments

Working from home? That will be extra. Renters rage over new fee

usatoday.com · 15 points · 7 comments

AMD Posts Linux Patches for HDMI 2.1 Auto Low-Latency Mode "ALLM" & VRR

phoronix.com · 10 points · 0 comments

An LLM-assisted security review of GlobaLeaks: 41 findings for $3,140

isgroup.biz · 8 points · 5 comments

Noisegate – a differential-privacy gateway for untrusted AI agents

github.com · 19 points · 0 comments

LLM Routers Have Become a Service Category of Their Own

techstrong.ai · 9 points · 2 comments

Pgtestdb's template cloning approach to testing is fast

brandur.org · 20 points · 2 comments

I Forked MinIO Object Storage and Made It Run Faster

github.com · 5 points · 1 comments

Adversarial Code Obfuscation for Defending Against LLM-Based Analysis

arxiv.org · 3 points · 1 comments

Skytrace – Self-hosted 3D ADS-B viewer with receiver coverage domes

sky.luftaquila.io · 2 points · 0 comments

Go LLM SDK for streaming, tool-calling AI backends (plus frontend React lib)

github.com · 60 points · 16 comments

A minimalist proxy for your local LLM cluster (~1100 lines)

github.com · 2 points · 0 comments

For the cheap price of a permanent tatto, get access to job interview

20 points · 15 comments

How to get more coding productivity with LLMs

6 points · 1 comments

LLMs and Xfwl4

spurint.org · 5 points · 3 comments

The Productivity Mirage

frantic.im · 359 points · 153 comments

LLM Honeypot

llm2human.pages.dev · 387 points · 106 comments

Multi-LLM – review plans and code across 12 coding CLIs, many models

github.com · 3 points · 1 comments

VernLLM – The AI resilience layer for TypeScript

vernllm.vercel.app · 3 points · 1 comments