Alibaba releases Qwen3.6-27B under Apache 2.0

Released Qwen3.6-27B under Apache 2.0 — SWE-bench Verified 77.2%

Why it matters

A dense 27B beat its 397B MoE predecessor, and the center of gravity of truly open models moved down to the 30B class

Source — Primary sources first; secondary reports are labelled “Press”

Pressremio.aihttps://www.remio.ai/post/qwen3-6-open-source-model-beats-a-397b-giant-while-alibaba-quietly-closes-weights-on-its-flagshi

Added in Week 35, 2026 · See that week

Same week

DeepSeek-V4PaperDeepSeek releases V4-Pro (1.6T/49B) and V4-Flash (284B/13B), MIT, 1M contextOfficialCohere announces acquisition of and merger with Germany's Aleph AlphaPressGPT-5.5Official

Nearby on the same axis

Moonshot AI releases Kimi K2.6PressDeepSeek releases V4-Pro (1.6T/49B) and V4-Flash (284B/13B), MIT, 1M contextOfficialLG AI Research releases EXAONE 4.5 33BOfficialMeta Superintelligence Labs releases Muse SparkOfficial