Alibaba has unveiled Qwen3.6-Plus, its third proprietary AI model released within just a few days. The model is made available through the Alibaba Cloud Model Studio API and features a substantial context window of one million tokens.
According to the Qwen team, a primary focus for Qwen3.6-Plus has been on significantly enhancing its capabilities for agentic coding, encompassing both frontend development and the execution of complex code tasks.
In benchmarks published by Alibaba, Qwen3.6-Plus partially outperforms Anthropic's older flagship model, Claude 4.5 Opus, which was superseded by the more powerful 4.6 Opus in December 2025. It is worth noting that some of these performance measurements were conducted by Alibaba itself.
While Qwen3.6-Plus demonstrates superior performance compared to its predecessor, the Qwen 3.5 model, and in certain scenarios surpasses Claude 4.5 Opus, the Claude 4.6 Opus, released in December 2025, achieved a score of 65.4 percent on Terminal-Bench 2.0, placing it ahead of Qwen3.6-Plus.
Alibaba traditionally released its Qwen models as open source; however, the company has recently shifted its strategy. The latest Qwen3.5-Omni is also not freely available. This change reflects Alibaba's intention to generate more revenue from enterprise customers with its proprietary models, especially as its cloud division faces intense competition from ByteDance.
Bloomberg reports that Alibaba is targeting $100 billion in AI revenue over the next five years. Qwen3.6-Plus is slated for integration into the Qwen chatbot application and the company's new enterprise AI service named Wukong.