Meituan LongCat team releases VitaBench

2025-10-20 17:05:06

Meituan's LongCat team today officially released VitaBench (Versatile Interactive Tasks Benchmark), a large-scale intelligent agent evaluation benchmark that closely resembles real-life scenarios and targets complex problems. VitaBench, based on three high-frequency real-life scenarios: ordering takeout, dining in restaurants, and traveling, has constructed an interactive evaluation environment encompassing 66 tools and implemented comprehensive task design across these scenarios.

Meituan

Email Subscription

Newsletters and emails are now available! Delivered on time, every weekday, to keep you up to date with North American business news.

Weekly Highlights

                                    A Republican lawmaker is seeking to halt all federal funding to New York City following the election of Zohran Mamdani as mayor
                            2025-11-09

                                    Supreme Court Pause on SNAP Benefits: Market Ripples, Policy Shock & Ticker Watch
                            2025-11-12

                                    After $2,000 Payments, Remaining Revenue to Slash National Debt
                            2025-11-16

                                    Trump Administration’s 50-Year Mortgage Gamble: What It Means for Housing, Builders and the Options Market
                            2025-11-12

                                    Gen Z expects to inherit money and assets—but their boomer parents aren’t planning on leaving anything behind
                            2025-11-11