Alibaba's Qwen 3.5 397B-A17 Beats Its Larger Trillion-Parameter Model — at a Fraction of the Cost
6 Articles
6 Articles
Alibaba's Qwen 3.5 397B-A17 beats its larger trillion-parameter model — at a fraction of the cost
Alibaba dropped Qwen3.5 earlier this week, timed to coincide with the Lunar New Year, and the headline numbers alone are enough to make enterprise AI buyers stop and pay attention.The new flagship open-weight model — Qwen3.5-397B-A17B — packs 397 billion total parameters but activates only 17 billion per token. It is claiming benchmark wins against Alibaba's own previous flagship, Qwen3-Max, a model the company itself has acknowledged exceeded o…
Can Alibaba’s Qwen3.5 Help Chinese AI Catch Up?
Alibaba’s answer to ChatGPT, Qwen, got an upgrade this week that could put it ahead in the AI race. The Chinese company said Qwen3.5 is cheaper and more powerful than its predecessor and performs on par with US rivals including Claude and Gemini. It can understand text, images and videos, and it’s compatible with trendy AI agents like OpenClaw. Qwen isn’t a standalone offering for Alibaba. Instead, it’s integrated into its main e-commerce biz. …
The latest version of Alibaba's Qwen is designed to support AI agent tasks and can include text, photo and video entries, the company reported in a statement on Monday. It can also analyze videos of up to two hours duration. Read more
Alibaba Unveils Qwen-3.5, Escalating Global AI Model Competition
Alibaba Cloud has unveiled Qwen-3.5, its next-generation open artificial intelligence model, which features multimodal capabilities and open weights. This release signals Alibaba’s ambition to anchor the next phase of global AI deployment, sharpening the global AI race between China and the US.
Alibaba Qwen 3.5 release dominates influencer discussions as open-source performance matches frontier models, reveals GlobalData
Alibaba Group Holding Ltd became a focal point for influencers in mid-Feb 2026, following the launch of its next-generation Qwen-3.5 AI series. The development features a flagship open-source model containing 397 billion parameters that utilize efficient architecture with 17 billion active parameters. The release stands out for introducing native multimodal capabilities and supporting a context window of one million tokens in its hosted version.…
Coverage Details
Bias Distribution
- 100% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium




