Claude vs Claude Code
People often think Claude Code is a model. Claude is the model, the part that reads and writes text. Claude Code is a program built around it. It gives Claude hands: it reads your files, runs commands and edits code on your computer. The same Claude models also run in the Claude app and in other tools. Claude Code is the one made for coding work.
Claude
The AI models made by Anthropic, in sizes from fast to strong.
Claude is Anthropic's family of large language models. It first came out in March 2023. It comes in sizes with names such as Haiku, Sonnet and Opus. The small ones are faster and cheaper, and the large ones are better at hard tasks. Claude can read long documents, look at images, write code and use tools. You reach it through the Claude app, Claude Code, Anthropic's API, and big clouds such as Amazon Bedrock and Google Cloud.
Claude Code
A coding agent from Anthropic. It reads, runs and edits the code in your project.
Claude Code is Anthropic's coding agent. You tell it what you want in plain words, and it reads your files, runs commands and edits your code. It works in a loop: it looks, acts, checks the result, then goes again until the job is done. It runs in a terminal, in code editors, in a desktop app and on the web. Claude is the model that thinks. Claude Code is the program around it that gives it hands.
In one line
- Claude
- The AI models made by Anthropic, in sizes from fast to strong.
- Claude Code
- A coding agent from Anthropic. It reads, runs and edits the code in your project.