AI/TLDR — New AI Releases Daily: Models, Tools, Repos & PapersA high-volume feed of new AI releases — models, open-source repos, developer tools, papers, datasets, and benchmarks — refreshed every 2 hours. Each release is explained in plain English so you actually understand what shipped.This site uses JavaScript to render the interactive feed. Enable JavaScript, or visit the source repo for the raw JSON.

AI/TLDR

SYSTEM 01/14 · THE FIELD GUIDE

LLM Fundamentals

How large language models actually work — tokens, transformers, context windows, and why they make things up.

6 TRACKS52 ARTICLESbeginner → advanced

LLM Basics

The core ideas behind every AI model you'll ever call.

Large Language ModelsBEGINNER LLM InternalsBEGINNER GPUs & AI ComputeINTERMEDIATE Scaling LawsINTERMEDIATE Inside ChatGPTBEGINNER Base vs Instruct vs ChatBEGINNER LLM vs GenAI vs AGIBEGINNER Foundation ModelsBEGINNER Emergent AbilitiesINTERMEDIATE Parameters & WeightsBEGINNER

Tokens & Tokenization

The unit everything is priced, limited, and measured in.

TokensBEGINNER Tokens vs Words vs CharsBEGINNER Byte-Pair EncodingINTERMEDIATE The Strawberry ProblemINTERMEDIATE Counting TokensBEGINNER Chat Templates & Special TokensINTERMEDIATE Tokenizing Other LanguagesINTERMEDIATE Vocabulary SizeINTERMEDIATE Unknown & Rare WordsINTERMEDIATE

Transformers & Attention

The architecture behind every modern model, without the math degree.

Transformer ModelsBEGINNER The Attention MechanismBEGINNER Mixture-of-Experts (MoE)ADVANCED FlashAttentionADVANCED Pretraining vs Post-TrainingINTERMEDIATE Encoder vs Decoder ModelsINTERMEDIATE Positional EncodingINTERMEDIATE Grouped-Query AttentionADVANCED Self vs Cross-AttentionADVANCED Residuals & Layer NormADVANCED

How Text Generation Works

From logits to the next word: how an LLM actually turns your prompt into text, one token at a time.

Next-Token PredictionBEGINNER Autoregressive GenerationINTERMEDIATE LogitsINTERMEDIATE The Softmax FunctionINTERMEDIATE Greedy Decoding vs SamplingINTERMEDIATE Speculative DecodingADVANCED Beam Search vs SamplingINTERMEDIATE Stop SequencesBEGINNER

Context Windows & Model Memory

What a model can hold in its head at once — and what happens when it can't.

Context WindowsBEGINNER Exceeding the WindowBEGINNER Lost in the MiddleINTERMEDIATE Million-Token WindowsADVANCED Context vs Training DataBEGINNER Measuring the WindowBEGINNER Sliding Window AttentionADVANCED

Sampling, Temperature & Hallucination

Why the same prompt gives different answers, and why some of them are wrong.

TemperatureBEGINNER HallucinationBEGINNER Top-p vs Top-kINTERMEDIATE Knowledge CutoffBEGINNER In-Context LearningINTERMEDIATE Frequency vs Presence PenaltyINTERMEDIATE Nondeterministic OutputsINTERMEDIATE LLMs and ArithmeticBEGINNER