Perplexity Launches WANDR to Test Research AI
Researching broad topics is difficult for AI. Developers need a way to measure if their models provide accurate facts and reliable sources. Perplexity created WANDR to solve this problem.
WANDR includes five hundred tasks that require deep searching and verification. It checks if an AI can find many relevant answers while citing real proof. This keeps developers honest about how well their systems perform.
Perplexity currently holds the top spot on this leaderboard. They hope that sharing this tool will help everyone build better and more accurate research agents.
Comments (0)
No comments yet. Be the first!
More AI news
NewsAlibaba releases powerful new Qwen 3.8 AI
Alibaba just launched its latest open-weight AI model which claims to outperform almost every competitor.
NewsAre video generators the secret to better computer vision?
Google Deepmind shows that video generators can handle complex visual tasks by learning from synthetic footage.
NewsKimi K3 Wins at Coding But Struggles With Math
Moonshot's new Kimi K3 model leads in web development tasks but falls short against top rivals in advanced mathematics.