Hugging Face

New Model additions VoxtralRealtime <img width="1920" height="1080" alt="image" src=" /> VoxtralRealtime is a streaming speech-to-text model from Mistral AI, designed for real-time a

How to train a new language model from scratch using Transformers and Tokenizers

New Model additions EXAONE-MoE <img width="2278" height="1142" alt="image" src=" /> K-EXAONE is a large-scale multilingual language model developed by LG AI Research. Built using a Mixture-of-Experts…

Transformers v5 release notes <img width="1800" height="1013" alt="image" src=" /> - Highlights - Significant API changes: dynamic weight loading, tokenization - Backwards Incompatible Changes - Bugfi…

Transformers v5 release notes <img width="1800" height="1013" alt="image" src=" /> - Highlights - Significant API changes: dynamic weight loading, tokenization - Backwards Incompatible Changes - Bugfi…

What's Changed Another fix for qwen vl models that prevented correctly loading the associated model type - this works together with of the previous patch release. * Fixed incorrect model_type for qwen…

What's Changed Should not have said last patch :wink: These should be the last remaining fixes that got lost in between patches and the transition to v5. QwenVL: add skipped keys in setattr as well by…

What's Changed Last patch release for v4: We have a few small fixes for remote generation methods (e.g. group beam search), vLLM, and an offline tokenizer fix (if it's already been cached). * Grouped…

Granite 4.0 Nano: Just how small can you go?

Swift Transformers Reaches 1.0 – and Looks to the Future

LeRobot v0.4.0: Supercharging OSS Robot Learning

Gradio 3.0 is Out!