Microsoft’s multi-agent AI system tops Anthropic’s Mythos on cybersecurity benchmark

Microsoft's new vulnerability-scanning system, codenamed MDASH, scored 88.45% on the CyberGym benchmark, surpassing ...

Microsoft MDASH Beats A Key Mythos Benchmark. Here’s Why That Matters

Microsoft MDASH outperforms Mythos Preview on the CyberGym benchmark, demonstrating improved vulnerability discovery ...
The Escapist

How To Use the Black Myth: Wukong Benchmark Tool

If you’d like to test your system and be sure it can run Black Myth: Wukong then here’s what you’ll need to do. We suggest you optimize your system first and you can start by choosing Benchmark from ...
VentureBeat

LiveBench is an open LLM benchmark that uses contamination-free test data and objective scoring

Want smarter insights in your inbox? Sign up for our weekly newsletters to get only what matters to enterprise AI, data, and security leaders. Subscribe Now A team of Abacus.AI, New York University, ...
The Next Web

OpenAI’s GPT-5.4 sets new records on professional benchmarks

The new model introduces native computer use, a 1-million-token context window, and a reworked tool-calling system. Whether it actually holds off Anthropic and Google is less clear. OpenAI is moving ...
MIT Technology Review

How to build a better AI benchmark

To fix the way we test and measure models, AI is learning tricks from social science. It’s not easy being one of Silicon Valley’s favorite benchmarks. SWE-Bench (pronounced “swee bench”) launched in ...