v5.8.0 is a minor release. Key changes include:

• Add DeepSeek V4 (#45643) by @ArthurZucker in #45643;
• First model (#45788) by @SindhuRaghuram97 in #45788.

v5.7.0 is a minor release. Key changes include:

• Laguna XS.2 implementation (#45673) by @joerowell in #45673;
• model: Add DEIMv2 to Transformers (#44339) by @harshaljanjani in #44339.

v5.6.2 is a patch release: fix configuration reading and error handling for kernels (https://github.com/huggingface/transformers/pull/45610) by @hmellor

v5.6.1 is a patch release: fix attributeerror on s_aux=none in flash_attention_forward (https://github.com/huggingface/transformers/pull/45589) by @jamesbraza

v5.6.0 is a minor release. Key changes include:

• [Privacy Filter] Add model (#45580) by @vasqu in #45580;
• add Qianfan-OCR model definition (#45280) by @marvinzh in #45280.

v5.5.4 is a patch release. Key changes include:

• Patch release v5.5.4;
• This is mostly some fixes that are good to have asap, mostly for tokenizers; ** Fix Kimi-K2.5 tokenizer regression and _patch_mistral_regex Attribute… (#45305)….

v5.5.3 is a patch release: [gemma4] fix device map auto (#45347) by @cyrilvallez

v5.5.2 is a patch release. Key changes include:

• Add MoE to Gemma4 TP plan (#45219) by @sywangyi and @Cyrilvallez;
• [gemma4] Dissociate kv states sharing from the Cache (#45312) by @Cyrilvallez.

v5.5.1 is a patch release. Key changes include:

• Patch release v5.5.1;
• This patch is very small and focuses on vLLM and Gemma4!.

v5.5.0 is a minor release. Key changes include:

• The total number of pixels must fit within a patch budget;
• Both height and width must be divisible by 48 (= patch size 16 × pooling kernel 3).

v5.4.0 is a minor release. Key changes include:

• Add VidEoMT (#44285) by @NielsRogge in #44285;
• [Model] Add UVDoc Model Support (#43385) by @XingweiDeng in #43385.

v5.3.0 is a minor release. Key changes include:

• Add eurobert (#39455) by @ArthurZucker in #39455;
• Add VibeVoice ASR (#43625) by @ebezzam in #43625.

v5.2.0 is a minor release. Key changes include:

• Add Voxtral Realtime (#43769) by @eustlb;
• Add GlmMoeDsa (#43858) by @Cyrilvallez.

v5.1.0 is a minor release. Key changes include:

• Add EXAONE-MoE implementations (#43080) by @nuxlear;
• [Model] Add PP-DocLayoutV3 Model Support (#43098) by @zhang-prog.

v5.0.0 is a major release. Key changes include:

• Highlights;
• Significant API changes: dynamic weight loading, tokenization.

v5.0.0rc3 is a unknown release. Key changes include:

• [GLM-4.7] GLM-Lite Supoort by @zRzRzRzRzRzRzR in https://github.com/huggingface/transformers/pull/43031;
• [GLM-Image] AR Model Support for GLM-Image by @zRzRzRzRzRzRzR in https://github.com/huggingface/transformers/pull/43100.

v4.57.6 is a patch release: fixed incorrect model_type for qwen2vl and qwen2.5vl when config is saved and loaded again by @i3hz in https://github.com/huggingface/transformers/pull/41758

v4.57.5 is a patch release. Key changes include:

• QwenVL: add skipped keys in setattr as well by @zucchini-nlp in https://github.com/huggingface/transformers/pull/41808;
• Fix lr_scheduler_parsing by @SunMarc in https://github.com/huggingface/transformers/pull/41322.

v4.57.4 is a patch release. Key changes include:

• Grouped beam search from config params by @zucchini-nlp in https://github.com/huggingface/transformers/pull/42472;
• Handle decorator with optional arguments better @hmellor in https://github.com/huggingface/transformers/pull/42512.

v5.0.0rc2 is a unknown release. Key changes include:

• batched and grouped experts implementations by @IlyasMoutawwakil in https://github.com/huggingface/transformers/pull/42697;
• Optimize MoEs for decoding using batched_mm by @IlyasMoutawwakil in https://github.com/huggingface/transformers/pull/43126.

v5.0.0rc1 is a unknown release. Key changes include:

• Default auto 🚨 🚨 by @ArthurZucker in https://github.com/huggingface/transformers/pull/42805;
• 🚨🚨 [saving] Default to 50GB shards, and remove non-safe serialization by @Cyrilvallez in https://github.com/huggingface/transformers/pull/42734.

v5.8.0 is a minor release. Key changes include:

• Add DeepSeek V4 (#45643) by @ArthurZucker in #45643;
• First model (#45788) by @SindhuRaghuram97 in #45788.

v5.7.0 is a minor release. Key changes include:

• Laguna XS.2 implementation (#45673) by @joerowell in #45673;
• model: Add DEIMv2 to Transformers (#44339) by @harshaljanjani in #44339.

v5.6.2 is a patch release: fix configuration reading and error handling for kernels (https://github.com/huggingface/transformers/pull/45610) by @hmellor

v5.6.1 is a patch release: fix attributeerror on s_aux=none in flash_attention_forward (https://github.com/huggingface/transformers/pull/45589) by @jamesbraza

v5.6.0 is a minor release. Key changes include:

• [Privacy Filter] Add model (#45580) by @vasqu in #45580;
• add Qianfan-OCR model definition (#45280) by @marvinzh in #45280.

v5.5.4 is a patch release. Key changes include:

• Patch release v5.5.4;
• This is mostly some fixes that are good to have asap, mostly for tokenizers; ** Fix Kimi-K2.5 tokenizer regression and _patch_mistral_regex Attribute… (#45305)….

v5.5.3 is a patch release: [gemma4] fix device map auto (#45347) by @cyrilvallez

v5.5.2 is a patch release. Key changes include:

• Add MoE to Gemma4 TP plan (#45219) by @sywangyi and @Cyrilvallez;
• [gemma4] Dissociate kv states sharing from the Cache (#45312) by @Cyrilvallez.

v5.5.1 is a patch release. Key changes include:

• Patch release v5.5.1;
• This patch is very small and focuses on vLLM and Gemma4!.

v5.5.0 is a minor release. Key changes include:

• The total number of pixels must fit within a patch budget;
• Both height and width must be divisible by 48 (= patch size 16 × pooling kernel 3).

v5.4.0 is a minor release. Key changes include:

• Add VidEoMT (#44285) by @NielsRogge in #44285;
• [Model] Add UVDoc Model Support (#43385) by @XingweiDeng in #43385.

v5.3.0 is a minor release. Key changes include:

• Add eurobert (#39455) by @ArthurZucker in #39455;
• Add VibeVoice ASR (#43625) by @ebezzam in #43625.

v5.2.0 is a minor release. Key changes include:

• Add Voxtral Realtime (#43769) by @eustlb;
• Add GlmMoeDsa (#43858) by @Cyrilvallez.

v5.1.0 is a minor release. Key changes include:

• Add EXAONE-MoE implementations (#43080) by @nuxlear;
• [Model] Add PP-DocLayoutV3 Model Support (#43098) by @zhang-prog.

v5.0.0 is a major release. Key changes include:

• Highlights;
• Significant API changes: dynamic weight loading, tokenization.

v5.0.0rc3 is a unknown release. Key changes include:

• [GLM-4.7] GLM-Lite Supoort by @zRzRzRzRzRzRzR in https://github.com/huggingface/transformers/pull/43031;
• [GLM-Image] AR Model Support for GLM-Image by @zRzRzRzRzRzRzR in https://github.com/huggingface/transformers/pull/43100.

v4.57.6 is a patch release: fixed incorrect model_type for qwen2vl and qwen2.5vl when config is saved and loaded again by @i3hz in https://github.com/huggingface/transformers/pull/41758

v4.57.5 is a patch release. Key changes include:

• QwenVL: add skipped keys in setattr as well by @zucchini-nlp in https://github.com/huggingface/transformers/pull/41808;
• Fix lr_scheduler_parsing by @SunMarc in https://github.com/huggingface/transformers/pull/41322.

v4.57.4 is a patch release. Key changes include:

• Grouped beam search from config params by @zucchini-nlp in https://github.com/huggingface/transformers/pull/42472;
• Handle decorator with optional arguments better @hmellor in https://github.com/huggingface/transformers/pull/42512.

v5.0.0rc2 is a unknown release. Key changes include:

• batched and grouped experts implementations by @IlyasMoutawwakil in https://github.com/huggingface/transformers/pull/42697;
• Optimize MoEs for decoding using batched_mm by @IlyasMoutawwakil in https://github.com/huggingface/transformers/pull/43126.

v5.0.0rc1 is a unknown release. Key changes include:

• Default auto 🚨 🚨 by @ArthurZucker in https://github.com/huggingface/transformers/pull/42805;
• 🚨🚨 [saving] Default to 50GB shards, and remove non-safe serialization by @Cyrilvallez in https://github.com/huggingface/transformers/pull/42734.