AI Model Hubs

One page per model with what it's actually good at, where it falls short, the tutorials worth reading, and the workflows we've tested with it.

ChatGPT

The default assistant for millions — and the model most workflows still ship on.

The long-context, tool-safe assistant that quietly runs a lot of production AI.

Google's multimodal model — cheapest at high volume and best when you live inside Workspace.

The X-native model with real-time context and a distinctly unfiltered personality.

Not a chatbot — an answer engine. The cleanest interface for citation-first AI research.

The AI-first IDE. Not autocomplete with an LLM — a fundamentally different way to write code.

Open-weights reasoning models with pricing that changes what production AI economics look like.

The autocomplete every developer already has — and increasingly a proper agent inside your IDE.

The default AI API — best-in-class function calling, structured outputs, and the widest ecosystem.

Claude in production — long-context leader, best prompt-following, and the reference tool-use ergonomics.

The million-token context leader, best-in-class native multimodality, and the deepest Google ecosystem tie-ins.

More hubs (Grok, Perplexity, DeepSeek, Cursor, Copilot, OpenAI API, Anthropic API, Google AI) publish as we complete their full write-ups.