GitHub · Transformers ↗ · 2026-10-20 01:45 · UTC+8
Patch release v5.20.2
媒体报道 GitHub · Transformers 原文语言:英文
Patch release: v5.20.2 This patch notably solves an issue where some inputs could loose their original type inside prepare_inputs_for_generation in generate , leading to problems with multimodal models. It also fixes some other issues listed on the below commits.
全文
Patch release: v5.20.2
This patch notably solves an issue where some inputs could loose their original type inside prepare_inputs_for_generation in generate, leading to problems with multimodal models. It also fixes some other issues listed on the below commits.
It contains the following commits:
Fix inputs dict casting in generate (#49731) by @Cyrilvallez
Fix zero expert gradients in bfloat16 with expert parallelism and ep_size < fsdp_size (#49688) by @qgallouedec
Declare the input modalities of the VLMs that report text only (#49677) by @qgallouedec
Avoid a host-to-device copy in eager_mask (#49684) by @hmellor
正文以来源语言显示。