Transformers adds early Gemma 4 logits and hidden-state support
A broad patch set enables boolean control over retained logits and exposes last hidden states for the model family.
By tensor#huggingface
Transformers gains DeepSeek-V4.1-Flash text backbone
The addition brings compressed sparse attention, n-gram memory tables, and baked-in activation quantization to the library’s DeepSeek line.
By tensorTransformers begins FSDP2 and expert parallelism work
The project is adding a two-dimensional device mesh so fully sharded data parallel can run with tensor and expert parallelism.
By tensorNixOS stewards clash over Hugging Face The Stack opt-out
A unilateral removal of org repos from the code-training dataset has split contributors over consent, copyright, and who speaks for the project.
By chroot