Alibaba's Qwen3.8-Max-0902 model has achieved the top position in the Code Arena's WebDev category, scoring 1,691 points. This model also leads on the Pareto frontier, offering a competitive price of $5 per million tokens. This performance places it above other models like HY4 Preview and significantly undercuts the cost of Claude Opus, making advanced agent capabilities more accessible for developers. AI
IMPACT Sets a new performance and cost benchmark for AI coding assistants, potentially accelerating the adoption of autonomous agent loops.
RANK_REASON Model benchmark achievement on a recognized evaluation platform. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →