Transformers v5.12.0 发布：新增 MiniMax-M3-VL、PP-OCRv6 和 Parakeet-RNNT

AI 摘要

Hugging Face Transformers 库发布 v5.12.0 版本，新增了 MiniMax-M3-VL 视觉语言模型、PP-OCRv6 轻量级 OCR 系统和 Parakeet-RNNT 语音识别模型。该版本还包含多项 bug 修复和 CI 改进，并感谢了社区贡献者。

正文节选

# Release v5.12.0 ## New Model additions ### MiniMax-M3-VL <img width="886" height="583" alt="image" src="https://github.com/user-attachments/assets/ae9dd96f-6877-4531-a06b-a756686f24e5" /> MiniMax-M3-VL is the vision-language member of the MiniMax-M3 family that pairs a CLIP-style vision tower with 3D rotary position embeddings with the MiniMax-M3 text backbone. It uses a mixed dense/sparse Mixture-of-Experts decoder with SwiGLU-OAI gated experts and a lightning indexer for block-

发布时间：2026-06-12 22:39

抓取时间：2026-08-02 00:23

来源机构：Hugging Face