Term
LLM
Large Language Model — a transformer-based neural network trained on text (and increasingly images, audio, code) to predict the next token. Modern LLMs range from 7 B to 1 T+ parameters and serve as the engine behind coding agents and AI assistants.
Background
LLMs are the base technology powering vibe-coding tools. They are trained in two main stages: pretraining (next-token prediction on internet-scale text) and post-training (instruction tuning, RLHF, and safety alignment). As of 2026, frontier LLMs include Anthropic Claude (Opus, Sonnet, Haiku), OpenAI GPT-5 and o-series, Google Gemini 2.5, and open-weight Llama, Qwen, and DeepSeek. Coding-specialised models (DeepSeek-Coder, Codestral) trade general capability for code performance.