Tagged: Benchmark

109 articles

New benchmark shows AI still misses the human side of mental health care

PsyEval is a new benchmark that tests how large language models perform on mental health knowledge, diagnostic assessment, and emotional support tasks. Source: news-medical.net

Meta’s new AI research chief says agents are next big real-world milestone

UC Berkeley's RDI centre earlier this month introduced Agents' Last Exam, a new benchmark that tests how well AI agents perform. Source: amp.scmp.com

Claude AI Beats Human Robotics Teams 20x: Anthropic Marks Physical AI Turn

Claude AI robotics benchmark shows Opus 4.7 finishing physical robot programming in 9 minutes, against 181 minutes for AI-assisted human teams, in Anthropic... Source: techtimes.com

AI Agent Triggers Nuclear Strike After Getting Outmaneuvered in Civilization VI

A new benchmark designed to test strategic reasoning found an AI-controlled empire spent 50 turns developing nuclear weapons to stop a rival's cultural... Source: decrypt.co

Grok Imagine Video 1.5 Goes Live: xAI Tops AI Video Leaderboard at 86 Percent Below Sora

Grok Imagine Video 1.5 is now generally available as the top-ranked image-to-video AI generator, priced 86 percent below Sora 2 Pro at 4.20 dollars per... Source: techtimes.com

Riverflow Launches 2.5: The World's Highest-Rated AI Image Model

Riverflow 2.5 Pro tops Design Arena's global benchmark across Image, Graphic Design and Image Editing. Riverflow 2.5 Pro introduces custom scoring,... Source: guardonline.com

Beyond the benchmark: Advancing security at AI speed

Read how Microsoft Security has advanced its agentic vulnerability detection system, codename MDASH, integrating into real-world workflows across Windows,... Source: microsoft.com

AI Agent Failure Detection and Root Cause Analysis with Strands Evals

When your AI agent fails in production, knowing that it failed is only the beginning. The harder question is why it failed and what to fix. Source: aws.amazon.com

NVIDIA Achieves Leading Agentic Coding Performance on First Agentic AI Benchmark | NVIDIA Technical Blog

AI agents have fundamentally changed the complexity of inference workloads. Until now, the industry has struggled to define a standard for measuring how… Source: developer.nvidia.com

China just beat Nvidia on a global robotics ranking. Is a new tech war brewing?

Spirit AI says its foundation model for embodied intelligence is the first from China to top the RoboArena global leaderboard. Source: amp.scmp.com

How Will Smith Eating Spaghetti Became a Benchmark for AI Video Progress

From a cursed 2023 meme to a 2026 stress test, the Will Smith spaghetti videos continue to reveal how far AI video generation has come. Source: hackernoon.com

MiniMax M3 debuts, eclipsing GPT-5.5 and Gemini 3.1 Pro on key benchmark performance for just 5-10% of the cost

Big news in enterprise AI broke over the weekend as Chinese AI startup MiniMax released its highly anticipated M3 large language model on Sunday evening... Source: venturebeat.com

Bill Gurley Calls Anthropic 'Midwifing a Deity'

Venture capitalist Bill Gurley, a general partner at Benchmark Capital, said on the All-In Podcast that after spending a month researching **Anthropic** he... Source: letsdatascience.com

Mastering Agentic Techniques: AI Agent Evaluation

Evaluating an AI model and evaluating an AI agent are related—but they answer fundamentally different questions. A model benchmark tests the capability of... Source: developer.nvidia.com

LLM Evals Are Based on Vibes — I Built the Missing Layer That Decides What Ships

This article shows a full working implementation in pure Python, with real benchmark numbers. Most teams evaluate LLM responses by reading them and guessing... Source: towardsdatascience.com

EQT wins €5b deeptech mandate, Decart nears $4b, and Uber's Delivery Hero increase

Venture Capital. • Decart, a real-time GenAI research lab, raised $300m at nearly a $4b valuation. Radical Ventures led, joined by insiders Benchmark,... Source: axios.com

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 single-model systems from Anthropic and... Source: geekwire.com

Introducing Harvey’s Legal Agent Benchmark

Harvey's Legal Agent Benchmark is an open-source benchmark built to evaluate and improve agent capabilities for supporting legal work. Source: harvey.ai

Bret Taylor's Sierra raises nearly $1 billion months after last capital push

Sierra's $950 million Series E funding round was led by Tiger and Google's GV, with participation from Benchmark, Sequoia, Greenoaks and others. Source: cnbc.com

AgentClinic puts medical AI through a more realistic diagnostic test

A new benchmark shows that passing medical exams is not enough; clinical AI agents must gather information, handle uncertainty, use tools, interpret images,... Source: news-medical.net