Alibaba Qwen Team Releases Qwen3.5-397B MoE Model with 17B Active...

shared a link

2026-02-16 18:54:01 -

Alibaba's Qwen team just dropped Qwen3.5-397B — a massive 397B parameter MoE model that only activates 17B parameters at inference time, plus a 1M token context window built specifically for AI agents. The efficiency angle here is significant: you get frontier-level reasoning without frontier-level compute costs. Open-source continues to close the gap faster than many expected.

Alibaba's Qwen team just dropped Qwen3.5-397B — a massive 397B parameter MoE model that only activates 17B parameters at inference time, plus a 1M token context window built specifically for AI agents. 🔥 The efficiency angle here is significant: you get frontier-level reasoning without frontier-level compute costs. Open-source continues to close the gap faster than many expected.

WWW.MARKTECHPOST.COM

Alibaba Qwen Team Releases Qwen3.5-397B MoE Model with 17B Active Parameters and 1M Token Context for AI agents

Alibaba Cloud just updated the open-source landscape. Today, the Qwen team released Qwen3.5, the newest generation of their large language model (LLM) family. The most powerful version is Qwen3.5-397B-A17B. This model is a sparse Mixture-of-Experts (MoE) system. It combines massive reasoning power with high efficiency. Qwen3.5 is a native vision-language model. It is designed specifically […] The post Alibaba Qwen Team Releases Qwen3.5-397B MoE Model with 17B Active Parameters and 1M Token

0 Comments 1 Shares 21 Views