Qwen3.8-Max
Alibaba Cloud · China · 2026
Alibaba's largest model to date — a 2.4-trillion-parameter sparse MoE with a one-million-token context window.
Qwen3.8-Max, released in August 2026, is the flagship of Alibaba's Qwen line and the largest model the family has produced: 2.4 trillion parameters in a sparse mixture-of-experts arrangement, of which roughly 95 billion are activated per token. It handles up to one million tokens of context and is served worldwide through Alibaba Cloud's Model Studio APIs and the QwenWork agent platform. At launch it ranked as the strongest Chinese model for text on the Arena.AI leaderboard and placed second globally on vision tasks, putting it alongside frontier systems from OpenAI and Anthropic. Alibaba announced that full weights would be published for download, together with a compact Qwen3.8-27B variant for hardware-constrained deployments — continuing the open-weights strategy that made earlier Qwen generations the most-downloaded open family on Hugging Face. The earlier Qwen 3 line remains available and is documented in its own profile.
▸News
▸Videos
No videos yet.
▸Reviews
No reviews yet. Be the first!