Claude Opus 5.5 Max Strongly Tops the Agent Rankings, Domestic Multi-Model Models Stabilize in the Mid-to-Upper Ranks
The latest Arena AI Large Model Weekly Rankings have been announced, with major manufacturers competing in comprehensive, coding, agent, and multimedia categories. Anthropic's new model performed impressively, while domestic large models maintained a stable tier across various subcategories. In the agent rankings, Claude Opus 5.5 (Max) entered the ranking and immediately claimed the top position, with a task confirmation success rate of 17.29%, demonstrating an overwhelming advantage.