TurboQuant: Redefining AI efficiency with extreme compression

Run a 1T parameter model on a 32gb Mac by streaming tensors from NVMe

LLM Neuroanatomy II: Modern LLM Hacking and Hints of a Universal Language?

Epoch confirms GPT5.4 Pro solved a frontier math open problem

Chat GPT 5.2 cannot explain the German word "geschniegelt"

Autoresearch on an old research idea

iPhone 17 Pro Demonstrated Running a 400B LLM

Project Nomad – Knowledge That Never Goes Offline

Flash-MoE: Running a 397B Parameter Model on a Laptop

Why craft-lovers are losing their craft

We rewrote our Rust WASM Parser in TypeScript – and it got 3x Faster

Cursor Composer 2 is just Kimi K2.5 with RL

Anthropic takes legal action against OpenCode

No AI in Node.js Core

I turned Markdown into a protocol for generative UI

2% of ICML papers desk rejected because the authors used LLM in their reviews

Claw Compactor: compress LLM tokens 54% with zero dependencies

Launch an autonomous AI agent with sandboxed execution in 2 lines of code

GPT‑5.4 Mini and Nano

Grace Hopper's Revenge

Mistral Small 4

Language Model Teams as Distrbuted Systems

Speed at the cost of quality: Study of use of Cursor AI in open source projects (2025)

How I write software with LLMs

LLMs can be exhausting

LLM Architecture Gallery

Codegen is not productivity

The 100 hour gap between a vibecoded prototype and a working product

Tree Search Distillation for Language Models Using PPO

Allow me to get to know you, mistakes and all

AI didn't simplify software engineering: It just made bad engineering easier

Optimizing Content for Agents

1M context is now generally available for Opus 4.6 and Sonnet 4.6

Shall I implement it? No

Kotlin creator's new language: a formal way to talk to LLMs instead of English

Document poisoning in RAG systems: How attackers corrupt AI's sources

Are LLMs not getting better?

I don't use LLMs for programming

Reliable Software in the LLM Era

I'm glad the Anthropic fight is happening now

AMD Ryzen AI NPUs are finally useful under Linux for running LLMs

Launch HN: RunAnywhere (YC W26) – Faster AI Inference on Apple Silicon

I built a programming language using Claude Code

Debian decides not to decide on AI-generated contributions

Claude struggles to cope with ChatGPT exodus

I'm Not Consulting an LLM

SWE-CI: Evaluating Agent Capabilities in Maintaining Codebases via CI

How to run Qwen 3.5 locally

LLM Writing Tropes.md

Training students to prove they're not robots is pushing them to use more AI