You can now try Claude Code, Anthropic's new AI agent for coding
You can now try it.
2 min. read
Published on
Share this article
Improve this guide
Read our disclosure page to find out how can you help MSPoweruser sustain the editorial team Read more
Key notes
- Anthropic launched Claude Code, an AI agent for coding, currently in research preview.
- Claude Code can speed up tasks, completing in one step what typically takes 45 minutes.
- It’s part of the Claude 3.7 Sonnet launch, with both quick responses and expanded thinking modes.
AI this, AI that. It seems like every company on the face of the Earth is quick to jump on the AI hype train. That includes Anthropic, which had just launched its first AI agent for coding, the Claude Code.
Currently in a research preview, Claude Code lets you work with AI directly from the command line. It can do things like read and edit code, run tests, and commit changes to GitHub.
But the real advantage of Claude Code is its ability to speed up work. Anthropic says that the tool can finish tasks in a single step that would normally take developers 45 minutes or more.
This brings us to more competitors, as the race to the top of the food chain of AI agents for coding is heating up. There are a lot of tools in the market besides Claude Code, including GitHub Copilot, Code Assist from Google’s Gemini, and Cursor AI, to name a few.
Claude Code is part of the launch of Claude 3.7 Sonnet, Anthropic’s latest model that offers both quick responses and more detailed, step-by-step reasoning.
The model comes in two variants: the original mode and the “expanded thinking” mode, which makes the most out of its reasoning capabilities.
While Claude 3.7 Sonnet is available to all users, Claude Code is still in the research phase, and Anthropic is gathering feedback to improve it. Still, you can try Claude 3.7 Sonnet on Anthropic’s platforms, as well as third parties like Google’s Vertex AI and Amazon Bedrock.
User forum
0 messages