VitaBench

🌱VitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks 📃 Paper • 🌐 Website • 🏆 Leaderboard • 🛠️ Code • 🤗 Dataset 🔔 News [2026-01] Qwen3-Max-Thinking reported our Vita-Bench to evaluate and demonstrate its tool use capabilities (the averge score of 4 domains)！We invite the community to adopt Vita-Be

Papers using VitaBench (2)

EnvFactory: Scaling Tool-Use Agents via Executable Environments Synthesis and Robust RL2026AutoForge: Automated Environment Synthesis for Agentic Reinforcement Learning2025