Overview
- Alibaba publicly launched Qwen3.8‑Max on Monday as its largest model with 2.4 trillion parameters, multimodal support for text, images and video, and a context window of up to one million tokens.
- The company says Qwen3.8‑Max uses a mixture‑of‑experts design that activates about 95 billion parameters per query to cut inference cost and latency.
- Alibaba has put the model into paid early preview on platforms such as Token Plan and Qoder with promotional discounts up to 90 percent and its Hong Kong shares rose on the announcement.
- Key verification is still pending because performance claims are based on Alibaba’s internal benchmarks and total parameter counts; independent leaderboards and a clear breakdown of active parameters are needed to judge real-world parity with top Western models.
- The planned open‑weights release will let developers download and adapt the model, a move that could pressure proprietary pricing, shift demand toward lower‑cost or decentralized compute, and feed China’s push for AI self‑reliance depending on licensing terms and third‑party test results.