RepoJournal
Hugging Face

@huggingface

Transformers, Datasets, and the open AI-model layer

Pick a date

The Wire · Showcase

INTEL XPU JOINS KERNEL-BUILDER, TORCH 2.10 DEPRECATED

By RepoJournal · Filed · About Hugging Face

Hugging Face kernels now ships Intel XPU support via Triton, while dropping legacy Torch 2.10 from the test matrix.

The kernels team launched XPU-kernels, a new skill that packages Intel's Xe-Forge workflow into kernel-builder alongside existing CUDA and ROCm skills [1]. This targets Intel Battlemage / Arc Pro B70 hardware via the Intel XPU Backend for Triton, giving teams a third acceleration path for optimized kernel development. At the same time, nix-builder removed Torch 2.10 support [2] and updated ROCm test coverage, streamlining the CI matrix. Separately, kernels now guards against segfaults on free-threaded Python interpreters by disabling arch kernels when ABI3 isn't available [3]. Over in diffusers, the team consolidated docs by removing the pipeline examples section and migrating content to model docs [4], cutting duplication and maintenance burden. A new CI utility checks forward/call method docstrings for consistency [5]. The TRL team shipped Qwen3-VL training templates with generation markers for correct SFT masking [6], fixed tiny Qwen model config bugs [7], [8], and dropped unjustified test skips [9]. Finally, pytorch-image-models added reproducible validation with a --seed argument mirroring train.py [10], so validation runs produce consistent results.

Action items

References

  1. [1] Add xpu-kernels skill - Intel XPU Triton kernel development ↗ huggingface/kernels
  2. [2] nix-builder: remove Torch 2.10 support ↗ huggingface/kernels
  3. [3] Do not load arch kernels on free-threaded Python ↗ huggingface/kernels
  4. [4] [docs] remove pipeline examples section ↗ huggingface/diffusers
  5. [5] [WIP] chore: add utilities to check if call/forward methods are documented. ↗ huggingface/diffusers
  6. [6] Add Qwen3-VL training chat template with generation markers ↗ huggingface/trl
  7. [7] Fix tiny Qwen3-VL `deepstack_visual_indexes` and drop the test skip (#5779) huggingface/trl
  8. [8] Remove non-existent params from tiny Qwen2-VL model (#5795) huggingface/trl
  9. [9] Drop unjustified `model.visual.` skip in GRPO / RLOO Qwen2.5-VL tests (#5780) huggingface/trl
  10. [10] validate.py: add --seed for reproducible validation ↗ huggingface/pytorch-image-models

FAQ

What changed in Hugging Face on May 21, 2026?
Hugging Face kernels now ships Intel XPU support via Triton, while dropping legacy Torch 2.10 from the test matrix.
What should Hugging Face teams do about it?
Test XPU kernel builds on Battlemage hardware if you target Intel Arc • Drop Torch 2.10 from any custom kernel CI pipelines • Update Qwen3-VL fine-tuning templates if using return_assistant_tokens_mask=True
Which Hugging Face repositories shipped on May 21, 2026?
huggingface/kernels, huggingface/diffusers, huggingface/trl, huggingface/pytorch-image-models

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.