Alibaba Qwen3.8-2.4T-A95B open-sources massive 2.4T model

Alibaba's Qwen team today open-sourced Qwen3.8-2.4T-A95B, a 2.4 trillion-parameter large language model that rivals top-tier proprietary models. This release marks a significant milestone in open-source AI, pushing the boundaries of what community-driven models can achieve. The model features 95 billion active parameters per token—a technique known as Mixture-of-Experts (MoE)—which optimizes performance while reducing computational costs.

Benchmark results indicate Qwen3.8-2.4T-A95B performs comparably to or better than closed-source models like OpenAI's GPT-5.6 Sol, Anthropic's Fable 5, and Anthropic's Claude Opus 4.8 across key metrics. With native support for a 1 million token context window, it enables long-form document analysis, multi-turn conversations, and complex code generation without excessive prompt truncation. This open-source release democratizes access to state-of-the-art AI capabilities, allowing researchers, developers, and enterprises to fine-tune the model for specialized use cases.