Transformers 5.20.1 集成 Reflection AI 模型 Beam
Hugging Face 的 Transformers v5.20.1 发布说明称,本次更新加入 Reflection AI 的新模型 Beam,并称其是面向编程与智能体场景、注重 token 效率的前沿规模模型。说明还称,Beam 面向希望在自己硬件、基础设施或云上运行并定制强大模型的团队。这一补丁版本同时修复了旧版 PyTorch 上加载模型的回归问题,并包含 HunYuan-VL 批量生成时图像分配和 DeepGEMM Mega MoE 专家并行等修复。
发布说明把 Beam 定位为可自行部署的前沿规模编程与智能体模型,但没有提供基准测试或许可信息来佐证。
Release v5.20.1
Special surprise including the new model Beam from Reflection AI 🥳

Beam is a token-efficient, frontier-scale model focused on coding and agentic use cases. It is built for teams that want to run a powerful model on their own hardware, infrastructure, or cloud, and customize it to their specific needs.
Links: Documentation
- Fix small bug in auto mappings (#49664) by @yonigozlan in #49664
Some real patch fixes
We broke loading models on older torch versions 😢 should be fixed now. But here is the full list of fixes
- Fix HunYuan-VL image assignment during batched generation (#49557) by @Arnav66692
- Fix DeepGEMM Mega MoE under expert parallelism(#49428) by @IlyasMoutawwakil
- Fix from_pretrained on torch 2.7: require torch>=2.8 for distributed checkpointing (#49652) by @naveecode
- Make DtensorShardOperation work on torch 2.6 and 2.7 (#49657) by @hazelxue
Full Changelog: v5.20.0...v5.20.1