Launched Muse Glimmer, an open-weight, 30-billion-parameter model optimized for local, resident
agent workflows. Compared to leading models of the same size, Muse Glimmer performs excellently in critical agent use cases and benchmarks, and is designed to run entirely on consumer-grade hardware, such as Macs or PCs equipped with high-performance GPUs. Upholding our long-standing tradition of sharing foundational AI research, we release model weights under a lenient Apache 2.0 license. 🔗 Read the original article via AIHOT · https://aihot.virxact.com/items/cmsn2u3mt067ero8opidmwvz7
推出 Muse Glimmer,一款开放权重、300 亿参数的模型,专为本地、常驻运行的智能体工作流优化。
与同尺寸领先模型相比,Muse Glimmer 在关键智能体用例和基准测试中表现出色,并设计为完全在消费级硬件(如 Mac 或配备高性能 GPU 的 PC)上运行。
秉承我们长期分享基础 AI 研究的传统,我们以宽松的 Apache 2.0 许可证发布模型权重。
🔗 阅读原文 via AIHOT · https://aihot.virxact.com/items/cmsn2u3mt067ero8opidmwvz7
The readable text on this page was extracted from the public source and organized with attribution, publication time and the original link. Copyright remains with the original author and publisher.