Arena.ai has launched the Agent Arena Index with six focused sub-indexes plus an Alignment Index, built on millions of real-world tasks and traces from its global community. The benchmarks measure how well agents complete complex workflows, use tools and adapt to feedback, while the Alignment Index gauges fidelity to user intent.
Arena
Platform offering public leaderboards for models and agents, built on community crowdsourced comparisons and real-world task tra
Curated Intelligence
Arena.ai (@arena)T1·Agents
62 pts
Arena.ai (@arena)T1·Agents
58 pts
Arena.ai research intern and CMU PhD candidate Wayne Chi introduced Interaction-as-a-Judge, a paradigm in which computer-use agents navigate real applications and games to produce interaction traces evaluated by an LLM judge or a trained reward model. The framework spans preference evaluation drawn from real Code Arena battles and correctness evaluation based on requirements mined from user complaints in Agent Arena.