3 entries with this tag
One new research article published: comprehensive analysis of Alibaba's Qwen3.8-27B — a 27B dense, native vision-language model with hybrid Gated DeltaNet + Gated Attention architecture, delivering 73.0 on Terminal Bench 2.1, 61.7 on SWE-bench Pro (surpassing Opus 4.6 Max), 84.3 on OSWorld-Verified, and 90.0 on MathVision, all in a model that fits on a single consumer GPU under Apache 2.0.
On August 14, 2026, Alibaba's Qwen team released Qwen3.8-27B — a 27B dense, native vision-language model with hybrid Gated DeltaNet + Gated Attention architecture, flexible thinking control, and Apache 2.0 licensing. The model delivers 73.0 on Terminal Bench 2.1 (within 5 points of Opus 4.6 Max), 61.7 on SWE-bench Pro, 84.3 on OSWorld-Verified, and 90.0 on MathVision, all in a model that fits on a single consumer GPU. Covers architecture, text and vision benchmarks, deployment guidance, and strategic implications for the local AI landscape.
One new research article published: comprehensive analysis of Meta's Muse Glimmer 30B — a distilled local-first agentic model running on consumer hardware with Apache 2.0 licensing and DFlash speculative decoding.