Alibaba released Qwen3.8-Max on Monday, August 4, 2026, describing it as its largest and most capable AI model to date. The Chinese tech giant claims the model’s performance rivals top systems from US labs Anthropic and OpenAI, as well as domestic competitor Moonshot AI’s Kimi K3.
Alibaba’s own benchmark testing shows Qwen3.8-Max broadly matching — and in some cases exceeding — Anthropic’s flagship model, Fable 5. On the Arena.AI crowdsourced leaderboard, Qwen3.8-Max ranks behind only Fable 5 and three models in Anthropic’s Opus family for general text tasks. For frontend coding, only two Claude Opus models and Kimi K3 score higher; for visual analysis, only Fable 5 outperforms it.
The model has 2.4 trillion parameters, the numerical values a model learns during training to process data and recognize patterns. By comparison, Moonshot’s Kimi K3 has 2.8 trillion parameters, though most leading US labs — including OpenAI and Anthropic — do not disclose parameter counts for their top systems.
Alibaba said it will release the model’s weights next week, making Qwen3.8-Max an open-weight system. Open-weight models give developers significantly more control than proprietary products, though they are more restrictive than traditional open-source software. The release marks a return to open-weight publishing for Alibaba after the company briefly shifted toward proprietary releases for its more advanced models earlier in 2026.
The launch adds to a rapid sequence of Chinese AI releases. Kimi K3’s weights were released last week, and both ByteDance and MiniMax released new video generation models on the same Friday. Beijing has promoted the open-weight strategy as a way to expand China’s influence in global AI governance and drive adoption of domestic technology.
The release could intensify an ongoing debate in Washington and Silicon Valley over how to manage open-weight AI systems and maintain the US technological edge over China. US industry has largely rallied around preserving access to open-weight models, while closed-model providers OpenAI and Anthropic face separate scrutiny following reported cyberattacks carried out by their own AI agents.
Source: The Verge