Alibaba previews 2.4 trillion parameter Qwen3.8-Max model for open release
The Neuron ● Covered by 5 sources
Alibaba announced a preview of its Qwen3.8-Max model, which contains 2.4 trillion parameters and will be released as open-weight. The model represents a significant scale increase in Alibaba's Qwen lineup, positioning it as a major open-source AI model comparable to frontier closed models. The open release could expand access to large language models and increase competition in the open-source AI market.
Why it matters
Alibaba's Qwen team previewed Qwen3.8-Max with 2.4 trillion parameters and promised the full version will be released as open-weight software that developers can download and run themselves. The company claims it trails only Claude Fable 5 in performance but has not yet published independent benchmarks.
Also covered by
- MarkTechPost — Alibaba Previews Qwen3.8-Max, a 2.4 Trillion-Parameter Multimodal Model, Days After Moonshot’s Kimi K3 Open-Weight Launch
- The Neuron — Artificial Analysis Reports Kimi K3 Token Efficiency
- The Neuron — Moonshot's Kimi K3 Open Model Released
- MarkTechPost — Moonshot AI Releases Kimi K3: A 2.8 Trillion Parameter Open MoE Model With Kimi Delta Attention and 1M Context