Nvidia Nemotron 3.5 Lightning

Stealing Reasoning Traces from Proprietary LLM APIs

Claude moves bound of the Riemann Hypothesis from 41.6% to 67.2%

Learning more about Claude's mathematical capabilities

What's the best programming language for coding agents?

Exploring Claude/GPT Knowledge Cutoffs and Pre-Training Timelines

Humanising LLM Outputs Is Dumb

How I use LLMs to learn complex topics

ChatGPT starts blocking direct requests to copy an author's style

“Code was never the hard part” is an insult to all programmers

I won't read LLM authored fiction

What is a product?

Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users

LLMs won't break symmetric crypto

Born Against, or why hobby programming communities are against LLM usage

Reddit is introducing a new moderator: AI

Why Erdős Problems Are Falling to AI

Rust-lang/rust is adopting an LLM policy

Zero-Mem: Zero-Token Memory Operations for LLM Agents

It's not a fear of "AI communism"; it's a fear of competitive market capitalism

Why Large Language Models Fail at Tabular Prediction

LLMs reward expertise

Smaller, faster, safer: running Kimi and GLM at scale

AirLLM 70B inference with single 4GB GPU

Prevent cognitive debt by manually retyping LLM-generated code

Don't be a meat proxy

Running Kimi K3 on MI355X at Better Performance per Dollar Than B300

Karpathy’s Pelican

Charlie Stross – On the non-use of AI in my writing process

Everyone is building LLM routers, we deprecated ours

Is AI reasoning right for the wrong reasons?

Run Kimi K3 using 29 GB of RAM at 0.50 tok/s

The End of an Era

Google fixed more Chrome bugs in June than over the past two years, thanks to AI

"the very foundation of modern academia has been blown to bits"

Exploring the "Dario and Amanda" Prompt

I flagged two research papers for fake authors and both were accepted as orals

Agent Skill to Force Docs in ASD-STE100 Simplified Technical English

GCC steering committee announces AI policy

You can't solve computer use by ignoring the interface

Kuna: Decompiler Development in the Age of Coding Agents

LLM Honeypot

Launch HN: Tokenless (YC S26) – Automatic model switching to save money

AI in Linux

TokenTown: A visual way to understand how LLMs work

Handbook.md shows that long policy documents do not reliably govern agents

Kimi K3 Architecture Overview and Notes

Now is the time to give LLMs access to the ACM digital library

"Opus 5 is a really bad model"

Don't ask an LLM for a confidence score