NewsBite analysis · AI, Business, Cloud
MarkTechPost
Alibaba Previews Qwen3.8-Max, a 2.4 Trillion-Parameter Multimodal Model, Days After Moonshot's Kimi K3 Open-Weight Launch
On July 19, 2026, during the World AI Conference in Shanghai, Alibaba previewed its latest AI model, Qwen3.8-Max, boasting 2.4 trillion parameters. Positioned as the second-best model in performance, it promises advancements in coding, data analysis, and office workflow tasks. The unveiling comes closely after the launch of Moonshot's Kimi K3, emphasizing competitive dynamics in the AI sector. Despite its ambitious features, full performance benchmarks and specifications are yet to be finalized.
MonitorConfidence: HighImmediate horizon
Published 19 Jul 2026, 22:40 UTC4 min read
Visit original article ↗Lead image — MarkTechPost
Summary
TL;DR
- Qwen3.8-Max features 2.4 trillion parameters, focusing on multimodal capabilities.
- It's positioned as the second-best AI model currently, following Fable 5.
- The preview was announced at the World AI Conference in Shanghai.
- Qwen3.8 aims to surpass previous models in coding and data tasks.
- Open weights for the model are expected soon but not yet available.
Interesting facts
- The model is Alibaba's first multimodal AI exceeding 1 trillion parameters.
- Community reactions highlight both enthusiasm for open-weight models and skepticism over unverified claims.
- The model's launch timing is seen as a strategic response to competitive releases.