We've introduced a new text_editor tool in the Anthropic API. It's designed for apps where Claude works with text files. With the new tool, Claude can make targeted edits to specific portions of text. This reduces token consumption and latency, all while increasing accuracy.

We've just shipped a new feature in Claude Code: extended thinking. You can simply ask Claude to “think”, “think more”, or “think harder” and it’ll show its extended thinking process. This is all powered by our hybrid reasoning model, Claude 3.7 Sonnet.

Another batch of features for Claude Code! Up first: Vim mode. This gives you the familiar insert/command modes for editing your prompts in Claude Code. Turn it on by typing the slash command /vim. But that's not all:

As a fan of weird AI benchmarks, I like MCBench, where you vote on which LLM makes the best Minecraft build based on a prompt Also interesting how much leaderboards converge no matter what metric: Claude 3.7 & 3.5 and GPT-4.5 lead here, too. Suggests an underlying characteristic

Everyone who ever tried to get data from an academic publication sadly knows what this means...

The latest post is about a new method, the “think” tool, that can result in remarkable improvements in Claude’s agentic tool use ability: anthropic.com/engineering/cl…

It’s been a big week for Claude Code. We launched 8 exciting new features to help devs build faster and smarter. Here's a roundup of everything we released:

Super happy we are launching our engineering blog with this one on the Thinking tool.. anthropic.com/engineering/cl…

people love MCP and we are excited to add support across our products. available today in the agents SDK and support for chatgpt desktop app + responses api coming soon!

We're hosting our first developer conference in May! More details here: anthropic.com/news/Introduci…

One of the coolest feature of the Claude 4 family is the ability to use extended thinking in between tool calls. This is especially useful when Claude needs to revisit a strategy after receiving new information. Let me know how you like it!

Have you tried Claude 4 (Opus or Sonnet)? How do you like it? What should we improve in the next version?

Claude Opus 4 on ARC-AGI Semi Private Eval Base * ARC-AGI-1: 22.5%, $0.40/task * ARC-AGI-2: 1.3%, $0.63/task Thinking 16K * ARC-AGI-1: 35.7%, $1.25/task * ARC-AGI-2: 8.6%, $1.93/task Opus 4 sets new SOTA (8.6%) on ARC-AGI-2

"A very important reason for me to prefer Claude Code is that it uses the token allocation assigned (...) to my Claude Max (...). It was such a problem for me (...) to have to pay for a subscription to web interfaces and also agentic IDEs..." One subscription for all your needs

By joining our Anthropic for Startups Program, you'll be the first to know our next round of founder and developer activations! Next up: catch the startups team at NY Tech Week's Anthropic Founder Salon 6/3 👇 partiful.com/events/nUf8wan… We'll go deep on the launch of the new

LLMs turn out to be really helpful in interior design planning. I asked a model to pick wall and beanbag colors based on current state (pic 1) and visualize it (pic 2) and I followed the recommendation. The outcome is awesome! (pic 3)

Wow! My knowledge of AI just increased by 100x. Thank you Anthropic and Jo Zhu Kennedy for hosting an incredible event and your partnership with Tech Week and a16z . #nytechweek This panel was next level . Chris Gorgolewski Merrill Lutsky Jeff Barg Michael Gerstenhaber Pim de Witte

We're hiring for the Claude Code team! In particular, we're looking for a Systems Engineer (job-boards.greenhouse.io/anthropic/jobs…) and an Eng Manager (job-boards.greenhouse.io/anthropic/jobs…). The Claude Code team is having a blast, come join us :)

I'm making a list of all the non-coding things people are doing with Claude Code. What are you using Claude Code for?

Come join my team! job-boards.greenhouse.io/anthropic/jobs…