Alibaba and DeepSeek launch advanced AI models that drive down inference costs and scale capabilities. Alibaba introduces Qwen3.8-Max, a 2.4-trillion parameter mixture-of-experts model with 95 billion active parameters per request, while DeepSeek releases its V4-Flash model to compete on low-cost inference pricing. These releases intensify China's enterprise AI market competition by making massive model performance dramatically cheaper to deploy at scale.
Opening Kapyn…