The Model Context Protocol (MCP) is an open standard for connecting AI agents to external systems. Connecting agents to tools and data traditionally requires a custom integration for each pairing, creating...
Today we’re upgrading the GPT‑5 series with the release of:GPT‑5.1 Instant: our most-used model, now warmer, more intelligent, and better at following your instructions.GPT‑5.1 Thinking: our advanced...
By Ken Rockot, Member of the Technical Staff and Ben Goodger, Head of Engineering, ChatGPT AtlasLast week, we launched ChatGPT Atlas, a new way to browse the web with ChatGPT by your side. In addition to...
A note from the author Special thanks to my friends Matt Harding and Allie Brosh for their insight and inspiration when writing this comic. Netflix background art drawn by Megan...
Skip to main content Gemini API Gemini API...
First, congrats to the Moonshot AI team, one of the 6 “AI Tigers” in China, on the awesome release of Kimi K2 Thinking. One of the overlooked and inspiring things for me these days is just how many people are...
K2 Vendor Verifier What's K2VV Since the release of the Kimi K2 model, we have received numerous feedback on the precision of Kimi K2 in toolcall. Given that K2 focuses on the agentic loop, the reliability of...
Our last post on Kimi K2 dives into how the Moonshot team used reinforcement learning (RL) on qualitative tasks. If you haven’t already, check out the last two explorations: How Rewriting Training Data...
Would you allow a stranger to drive a camera-equipped computer around your living room? You might have already done so without even realizing it. The Beginning: A Curious Experiment It all started innocently enough. I had recently bought an iLife A11 smart vacuum—a sleek, affordable, and technologically advanced robot promising effortless cleaning and intelligent navigation. As a curious engineer, I was fascinated by its workings. After leaving it to operate for the entire year, my curiosity
In Brief Posted: 9:55 AM PDT · August 2, 2025 Image Credits:Anthropic Anthropic has revoked OpenAI’s access to its Claude family of AI models, according to a report...
SWE-bench Bash Only uses the SWE-bench Verified dataset with the mini-SWE-agent environment for all models [Post]. SWE-bench Lite is a subset curated for less costly evaluation [Post]. ...
Bloomberg is reporting new details on the partnership between Apple and Google to have Gemini power the revamped Siri next year. Here are the details. 1.2 trillion paramenters for $1 billion A few days ago,...
A data-driven comparison of CSR, SSR, and RSC under the same app and test setup, focusing on initial-load performance and the impact of client- vs server-side data fetching (including Streaming + Suspense).
Today, we’re releasing two big updates that make Cursor the best place to work with agents: our first coding model, Composer, and a new interface for working with many agents in parallel.Introducing...
A Quiet Trend in the JavaScript Ecosystem For years, JavaScript has had exactly one meaningful directive, "use strict". It is standardized, enforced by runtimes, and behaves the same in every environm...