Alibaba's Qwen team is preparing to release Qwen3.8, a large language model with 2.4 trillion parameters, which is currently in a preview phase. The model is noted for its improving performance, particularly in web frontend tasks, and its token efficiency compared to other leading models. Alibaba plans to make Qwen3.8 open-weight soon, following its debut on various platforms including the Token Plan and Qoder. AI
IMPACT This release, particularly its open-weight nature and large parameter count, could significantly influence the open-source LLM landscape.
RANK_REASON Frontier-lab model release with system card.
AI-generated summary · Google Gemini · from 8 sources. How we write summaries →