Hacker News - Newest: "LLM" hnrss.org

Tamp: Cut LLM context size ~50% without changing your code

tamp.dev·5w·Hacker News

Mercury 2 on PinchBench: Diffusion LLM benchmarked on real OpenClaw agent tasks

inceptionlabs.ai·5w·Hacker News

a CLI tool to check if your local LLM setup is a security problem

iscooked.com·5w·Hacker News

Writing an LLM from scratch, part 32g – Interventions: weight tying

gilesthomas.com·6w·Hacker News, Hacker News

AgentBPF: eBPF-based observability for LLM agent trajectories

github.com·6w·Hacker News

Stop Engineering Prompts, Start Engineering Context: A Guide to the “Agent Skills” Standard

medium.com·6w·Hacker News

IvY-Rsearch/precomit: Measuring how LLM outputs form before token commitment — a prompt-history effect on pre-commitment generation dynamics.

github.com·6w·Hacker News

Show HN: LLM prompts as CLI progs with args, piping, and SSH forwarding

news.ycombinator.com·6w·Hacker News

Xiaomi stuns with new MiMo-V2-Pro LLM nearing GPT-5.2, Opus 4.6 performance at a fraction of the cost

venturebeat.com·6w·Hacker News

Autoresearching Apple’s “LLM in a Flash” to run Qwen 397B locally

simonwillison.net·6w·Hacker News, Hacker News

Observability for your LLM-powered apps: OTel Instrumentation for RubyLLM

thoughtbot.com·7w·Hacker News

maxmetcalfe/rede: Cloudflare Workers + Durable Objects runtime for small networks of LLM-backed bots.

github.com·6w·Hacker News

On Violations of LLM Review Policies

blog.icml.cc·6w·Hacker News, Hacker News

Show HN: ATO – a GUI to see and fix what your LLM agents configured

github.com·6w·Hacker News

Show HN: A Genetic algorithm that red-teams your copy with 100 LLM personas

crashtestcopy.com·6w·Hacker News

Show HN: Duplicate 3 layers in a 24B LLM, logical deduction .22→.76. No training

github.com·6w·Hacker News

anilatambharii/argus-ai: The world's first AI observability platform that doesn't just alert you- it fixes itself. Most tools stops at showing you the problem. ARGUS closes the loop autonomously. I built the self-healing AI ops platform that closes the loop monitoring tools never could. Production-grade LLM observability in 3 lines of Python. G-ARVIS scoring

github.com·6w·Hacker News, Hacker News

xybrid-ai/xybrid: On-device AI runtime for Flutter, Unity, iOS & Android. LLMs, TTS, ASR — no cloud required.

github.com·6w·Hacker News

Characterizing Delusional Spirals Through Human-LLM Chat Logs

spirals.stanford.edu·6w·Hacker News

570,000 Lines of LLM Code Compiled Fine. It Was 20,171x Slower Than SQLite.

tonylee.im·6w·Hacker News

Log in to enable infinite scrolling