huggingface/transformers v5.20.2: Patch release v5.20.2
retained text
# 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
document fetched 2026-10-24 02:34