EasyHubExplore
Explore
EasyHubTools & knowledge
All toolsEasyHub Blog
EN

Site appearance

Your color. Your style.

Accent colorRose
Visual styleSame content, fresh look

Soft gradients, dimensional icons

Applied: Rose · Studio. Saved in this browser.

EasyHubExplore
Home中文

Development·EASYHUB JOURNAL

Transformers 5.19.0 adds EmbeddingGemma 2 and expands MoE expert parallelism and continuous batching

Hugging FaceSource published
Editorial illustration: Transformers 5.19.0 adds EmbeddingGemma 2 and expands MoE expert parallelism and continuous batching
EasyHub editorial illustration · Not a source photograph

What changed

Hugging Face released Transformers 5.19.0 on October 6. The release adds EmbeddingGemma 2 support and substantial work on MoE expert parallelism, token dispatch, Trainer compatibility and continuous batching, along with cache, attention and model-compatibility fixes. It also includes breaking changes around MoE router-logit outputs and deprecation of paged attention prefixes.

Read original source
Original title
Release v5.19.0
Source
Hugging Face · github.com
Topic
Development
Source month
2026-10

This is a concise EasyHub summary of the linked source, not the full report or original reporting. Availability and preview conditions are described in the summary and original.

Summary page published · Editorial information

Back to the news timeline