Quasar-10B Brings Linear Attention to 2M-Token AI Contexts
aimodels44
·
2026-06-11
·
via HackerNoon - programming
Quasar-10B is a 10B foundation model using Gated Linear Attention to process 2M-token contexts with linear co…
此内容由惯性聚合(RSS阅读器)自动聚合整理,仅供阅读参考。 原文来自 — 版权归原作者所有。