{"version":"https://jsonfeed.org/version/1.1","title":"Transformers — Lab Ledger Desk","home_page_url":"https://labledgerdesk.pages.dev/lab/transformers/","feed_url":"https://labledgerdesk.pages.dev/lab/transformers/feed.json","description":"Open-source Transformers releases filed by Lab Ledger Desk.","items":[{"id":"https://labledgerdesk.pages.dev/b/2026/11/13/release-v5-23-0/","url":"https://labledgerdesk.pages.dev/b/2026/11/13/release-v5-23-0/","title":"Release v5.23.0","content_text":"Release v5.23.0 New Model additions Inkling2 Inkling 2 Small is a multimodal model from Thinking Machines Lab that accepts text and image inputs and generates text outputs, and it is intended for use in English and other languages and across multiple coding languages. …","date_published":"2026-11-13T14:10:57.000Z","authors":[{"name":"Transformers"}],"tags":["launch","llm"],"external_url":"https://github.com/huggingface/transformers/releases/tag/v5.23.0"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/06/transformers-v5-22-2/","url":"https://labledgerdesk.pages.dev/b/2026/11/06/transformers-v5-22-2/","title":"Patch release v5.22.2","content_text":"Patch release v5.22.2 Some in between patch for security and distributed fixes. See the included commits 🤗 [Distributed] Set device_id only for NCCL ( #50281 ) by @michaelbenayoun and @3outeille Harden our Jinja sandboxing ( #50431 ) by @Rocketknight1 Full Changelog : v5.22.1...v5.22.2","date_published":"2026-11-06T15:39:22.000Z","authors":[{"name":"Transformers"}],"tags":["launch"],"external_url":"https://github.com/huggingface/transformers/releases/tag/v5.22.2"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/30/transformers-v5-22-1/","url":"https://labledgerdesk.pages.dev/b/2026/10/30/transformers-v5-22-1/","title":"Patch release v5.22.1","content_text":"Patch release v5.22.1 We just had to fix the documentation! ** [Fix] Remove WIP doc for Ml4 ( #50124 ) from @remi-or","date_published":"2026-10-30T07:20:08.000Z","authors":[{"name":"Transformers"}],"tags":["launch"],"external_url":"https://github.com/huggingface/transformers/releases/tag/v5.22.1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/29/v5-22-0-mistral-large-4-le-chonk/","url":"https://labledgerdesk.pages.dev/b/2026/10/29/v5-22-0-mistral-large-4-le-chonk/","title":"v5.22.0: Mistral Large 4 (Le Chonk)","content_text":"New Models Mistral Large 4 (Le Chonk) Mistral Large 4 is Mistral's largest and most capable model to date: a 1 trillion-parameter natively multimodal Mixture-of-Experts model with 49 billion active parameters per token (52 billion including embeddings and output layers), trained from scratch on 3,800 NVIDIA Grace…","date_published":"2026-10-29T14:16:58.000Z","authors":[{"name":"Transformers"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/huggingface/transformers/releases/tag/v5.22.0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/29/transformers-v5-21-1/","url":"https://labledgerdesk.pages.dev/b/2026/10/29/transformers-v5-21-1/","title":"Patch release v5.21.1","content_text":"Patch release v5.21.1 ** Fix rope_scaling setter for configs that assign it during init ( #50080 ) We broke remote code loading in vllm, this fixes it sorry everyone!","date_published":"2026-10-29T07:56:35.000Z","authors":[{"name":"Transformers"}],"tags":["launch"],"external_url":"https://github.com/huggingface/transformers/releases/tag/v5.21.1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/26/transformers-v5-21/","url":"https://labledgerdesk.pages.dev/b/2026/10/26/transformers-v5-21/","title":"Release: v5.21.0","content_text":"Release v5.21.0 New Model additions MiniCPM-OCR MiniCPM-OCR is a document-understanding model from OpenBMB's MiniCPM-V family. It keeps the MiniCPM-V 4.6 architecture, which pairs a SigLIP vision encoder and a window-attention merger with a Qwen3.5 language model. …","date_published":"2026-10-26T17:47:07.000Z","authors":[{"name":"Transformers"}],"tags":["launch","llm"],"external_url":"https://github.com/huggingface/transformers/releases/tag/v5.21"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/19/patch-release-v5-20-2/","url":"https://labledgerdesk.pages.dev/b/2026/10/19/patch-release-v5-20-2/","title":"Patch release v5.20.2","content_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. …","date_published":"2026-10-19T17:45:37.000Z","authors":[{"name":"Transformers"}],"tags":["launch"],"external_url":"https://github.com/huggingface/transformers/releases/tag/v5.20.2"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/16/transformers-v5-20-1/","url":"https://labledgerdesk.pages.dev/b/2026/10/16/transformers-v5-20-1/","title":"Release v5.20.1","content_text":"Release v5.20.1 Special surprise including the new model Beam from Reflection AI 🥳 Beam is a token-efficient, frontier-scale model focused on coding and agentic use cases. It is built for teams that want to run a powerful model on their own hardware, infrastructure, or cloud, and customize it to their specific needs. …","date_published":"2026-10-16T12:45:10.000Z","authors":[{"name":"Transformers"}],"tags":["launch"],"external_url":"https://github.com/huggingface/transformers/releases/tag/v5.20.1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/15/release-v5-20-0/","url":"https://labledgerdesk.pages.dev/b/2026/10/15/release-v5-20-0/","title":"Release v5.20.0","content_text":"Release v5.20.0 New Model additions Muse Spark Open Muse Spark Open is a multimodal mixture-of-experts model from Meta that takes text, images, audio, and video as input and generates text. It has 816B total parameters (42B active per token) and a context window of up to 1,048,576 tokens shared across modalities. …","date_published":"2026-10-15T18:05:36.000Z","authors":[{"name":"Transformers"}],"tags":["launch","llm","agents"],"external_url":"https://github.com/huggingface/transformers/releases/tag/v5.20.0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/06/transformers-v5-19-0/","url":"https://labledgerdesk.pages.dev/b/2026/10/06/transformers-v5-19-0/","title":"Release v5.19.0","content_text":"Release v5.19.0 New Model additions EmbeddingGemma2 EmbeddingGemma 2 is a multimodal embedding model from Google built on the Gemma 4 architecture. …","date_published":"2026-10-06T16:22:30.000Z","authors":[{"name":"Transformers"}],"tags":["launch"],"external_url":"https://github.com/huggingface/transformers/releases/tag/v5.19.0"}]}