Google DeepMind’s Gemini 4 Carbon (High) entered Agent Arena’s overall ranking at No. 2 on Oct. 28 after more than 21,900 sessions, posting a 16.2% net-improvement score with a median cost of $1.80 per task, 72% below the cost of the model in first place. Xiaomi’s MiMo V2.6 Pro led the open-weight category in mid-October with a 3.8% net-improvement score and costs that can fall to $0.11 per task; open-weight models make their trained parameters available so developers can run them on hardware they control. Anthropic’s Claude Opus 5.5 led overall performance at about 20% net improvement as of early November. Agent Arena’s frontier tracks the strongest available result at each price, and on the highest-cost tasks Google and Xiaomi models held frontier spots while OpenAI and Anthropic models slipped on that measure. Agent Arena rates models on jobs requiring several connected actions, including work with web search and file systems, with scores reflecting task completion and user feedback; since its June 2026 launch, the platform has gathered data from more than two million user sessions. A prior snapshot placed OpenAI’s GPT-6.1 Terra eighth overall with a 10.6% net-improvement score.