RepoJournal
Hugging Face

@huggingface

Transformers, Datasets, and the open AI-model layer

Pick a date

The Wire · Showcase

TRANSFORMERS FIXES MASK REGRESSION BEFORE IT BREAKS YOUR GENERATION PIPELINE

By RepoJournal · Filed · About Hugging Face

A critical mask return-type contract regression in transformers got patched overnight, along with fixes for offloaded model serialization that would have silently corrupted weight conversions in production.

The transformers team caught and fixed a mask handling regression [1] that broke the return-type contract for `create_masks_for_generate`, adding strict guards to prevent XPU causal mask skipping when kv_offset is non-zero. This is the kind of silent correctness bug that makes it past local tests. In parallel, they shipped a fix for `save_pretrained` [2] where offloaded weights on meta device were getting converted before reloading from disk, leaving converted tensors stranded on meta and breaking serialization workflows. Over in diffusers, the Flash-3 attention implementation got its own mask handling fix [3] for non-contiguous tensor support, addressing the same class of masking issues upstream. On the training side, GOLDTrainer now supports vision-language models [4] with both same-family distillation using JSD loss and cross-family distillation via separate image processor paths, meaning you can now train VLM students against teacher models without architectural alignment constraints. The hub-docs inference providers documentation regenerated automatically [5] as dependencies updated.

Action items

References

  1. [1] fix mask return-type contract regression and add correctness guard for ↗ huggingface/transformers
  2. [2] Fix save_pretrained with offloading and weight conversions ↗ huggingface/transformers
  3. [3] fix `_flash_3_varlen_hub` mask handling ↗ huggingface/diffusers
  4. [4] [GOLD] VLM support for GOLDTrainer ↗ huggingface/trl
  5. [5] [Bot] Update Inference Providers documentation ↗ huggingface/hub-docs

FAQ

What changed in Hugging Face on July 6, 2026?
A critical mask return-type contract regression in transformers got patched overnight, along with fixes for offloaded model serialization that would have silently corrupted weight conversions in production.
What should Hugging Face teams do about it?
Pin transformers to include the mask regression fix [ref:3] before your next generation deployment • Test model serialization workflows with offloaded weights after pulling the save_pretrained fix [ref:4] • Review GOLDTrainer VLM support [ref:9] if you're training vision-language model distillation
Which Hugging Face repositories shipped on July 6, 2026?
huggingface/transformers, huggingface/diffusers, huggingface/trl, huggingface/hub-docs

Related across the cluster

For your repos

The showcase is a teaser.
Your wire is the product.

Same engine. Different stack. Below: what changes when the wire is yours.

Showcase wire

  • 14 famous open source orgs
  • One wire per day
  • Public, generic
  • Read on the web, when you remember

Your wire

  • Up to 1,500 of your repos - orgs, deps, vendors
  • Morning and evening briefs
  • Action items routed to your team
  • Slack delivery, email, breaking-news CVE alerts

Want a hands-on demo first? Ask a current user for an invite link.