Transformers v5.15.0 发布:新增 Muse Glimmer 等多模型支持
原标题:Release: v5.15.0
AI 摘要
Hugging Face Transformers 发布 v5.15.0 版本,新增多个模型支持,包括 Meta 的 Muse Glimmer(30B 参数多模态模型,专为智能体场景设计,Apache 2.0 许可)、IBM 的 GraniteSWA/GraniteMoeSWA、SKT 的 A.X-K1/K2 和 Cosmos3 Edge。该版本还包含多项破坏性变更,如线性注意力模型内核改为可选、缓存裁剪 API 仅接受负值、T5 系列默认支持 SDPA 等。此外,版本修复了多项注意力机制和视觉模型的问题,并优化了性能。
正文节选
# Release v5.15.0 ## New Model additions ### Meta Muse Glimmer Muse Glimmer, released today, is Meta’s new multimodal model, especially designed for agentic use cases. Distilled from Muse to 30B parameters, and released under the Apache 2.0 license, it can be deployed to local setups for privacy-aware applications such as coding, document analysis, personal assistants, Claw- or Hermes-like setups. Muse Glimmer is a dense 30B parameter model consisting of: - 2B ViT-style encoder for