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 22 of 30

RSS feed for LLM

mlxsh A lightweight CLI to serve multiple local LLMs on Apple

github.com · 3 points · 1 comments

What's Next for LLMs?

4 points · 4 comments

Atomadic- Zero-LLM Sub-200us MCP Action Interlock (Storefront Restored)

atomadic.tech · 2 points · 0 comments

AI Mania: From Tulips to Tokens

seanhelvey.com · 49 points · 56 comments

KotlinLLM

github.com · 4 points · 1 comments

GPUs could explode to multiple TB with new storage-inspired memory tech

theregister.com · 32 points · 9 comments

CaLLMar – play a text-based adventure game in an LLM chat

github.com · 4 points · 0 comments

Running a 35B LLM at 128K Context, Full Speed, on €870 of Used Hardware

medium.com · 3 points · 1 comments

California to destroy 420000 peach trees. Del Monte closes canning facilities

ubirataonline.com.br · 6 points · 0 comments

JobRadar: Open-source job search agent that scores listings with a local LLM

github.com · 7 points · 0 comments

I get 25 deep researched ideas with one single prompt

github.com · 9 points · 2 comments

CostPerPrompt – Live AI API pricing and real-workload cost calculators

costperprompt.com · 20 points · 8 comments

Smevals: A small eval suite for evaluating models, prompts, and harnesses

simonwillison.net · 4 points · 1 comments

Symbio self fine-tuning AI loop

github.com · 10 points · 5 comments

Rails patches critical Active Storage flaw with RCE potential

bleepingcomputer.com · 5 points · 0 comments

Free AI Prompt Gen – A local-first, open-source prompt engineering tool

freeaipromptgen.com · 3 points · 0 comments

AllMCPs – Directory of MCP Servers

allmcps.com · 2 points · 0 comments

Pgtestdb's template cloning approach to testing is fast

brandur.org · 67 points · 51 comments

Cursor removed cost information from the usage page and CSV export

forum.cursor.com · 332 points · 153 comments

Aurora – AI Gateway built in Go

github.com · 7 points · 2 comments

Minimal LLM Post-Training Experiments on an 8GB GPU (SFT, DPO, GRPO)

github.com · 21 points · 0 comments

Horus-runtime – Train your own tiny LLM from scratch

github.com · 2 points · 0 comments

Lightweight, S3-compatible object storage server

vaults3.com · 4 points · 2 comments

LLMs can't trade and higher reasoning doesn't help

twitter.com · 5 points · 0 comments

US lawmakers investigate DoorDash's use of Moonshot AI's Kimi K2.6 model

scmp.com · 11 points · 4 comments

Sanitizer – Strip sensitive data from documents locally before an LLM

provexar.ai · 2 points · 0 comments

Yann LeCun on What Comes After LLMs [video]

youtube.com · 4 points · 0 comments

GAI – A Go runtime for typed, tool-using LLM agents

github.com · 3 points · 0 comments

Predictive Speculative KV Replication for Bursty LLM Inference

jwlabs.vercel.app · 41 points · 4 comments

Everyone is building LLM routers, we deprecated ours

manifest.build · 130 points · 86 comments