RepoJournal
Hugging Face

@huggingface

Transformers, Datasets, and the open AI-model layer

Pick a date

  1. Thu 28 may
  2. Fri 29 may
  3. Sat 30 may
  4. Sun 31 may
  5. Mon 1 jun
  6. Mon 8 jun
  7. Tue 9 jun
  8. Wed 10 jun
  9. Thu 11 jun
  10. Fri 12 jun
  11. Sat 13 jun
  12. Sun 14 jun
  13. Mon 15 jun
  14. Tue 16 jun
  15. Wed 17 jun
  16. Thu 18 jun
  17. Sat 27 jun
  18. Sun 28 jun
  19. Mon 29 jun
  20. Tue 30 jun
  21. Wed 1 jul
  22. Thu 2 jul
  23. Fri 3 jul
  24. Sat 4 jul
  25. Sun 5 jul
  26. Mon 6 jul
  27. Tue 7 jul
  28. Wed 8 jul
  29. Thu 9 jul
  30. Today 10 jul

The Wire ยท Showcase

TRANSFORMERS FIXES GPU IMAGE MELTDOWN, TRL GRADUATES KTO TO STABLE

By RepoJournal ยท Filed ยท About Hugging Face

The nightly all-latest-gpu image was broken by a dependency conflict. It's fixed. And KTO just left experimental.

The executorch dependency was uninstalling the pinned cu126 torch in the GPU image, pulling a mismatched CUDA/NCCL stack that crashed at import [1]. That's now removed from the GPU build, and executorch export tests have been reclassified as CPU-only to prevent future cross-stack breakage [2]. In parallel, transformers cleaned up pipeline generators to stop materializing them entirely, which was causing OOMs on complex inputs [3]. Over in TRL, version 1.8.0 promotes KTOTrainer from experimental to stable [4], and trainers now accept DatasetDict and IterableDatasetDict directly as eval_dataset, which means you can pass load_dataset output without reshaping [5]. On the diffusers side, Cosmos3 now has full modular pipeline support with dedicated encoding, denoising, and decoding blocks [6]. In smolagents, transformers 5.13.0 is now pinned out due to a compatibility issue that was breaking CI [7].

Action items

References

  1. [1] Remove executorch from all-latest-gpu image + add torch smoke test (#47196) huggingface/transformers
  2. [2] Make executorch exporter tests always use xnnpack backend (#47201) huggingface/transformers
  3. [3] Cleanup pipelines, stop materializing generators โ†— huggingface/transformers
  4. [4] v1.8.0 โ†— huggingface/trl
  5. [5] Support DatasetDict and IterableDatasetDict as eval_dataset in trainers โ†— huggingface/trl
  6. [6] Cosmos3 ModularPipeline โ†— huggingface/diffusers
  7. [7] Fix CI AttributeError: 'str' object has no attribute '__module__' โ†— huggingface/smolagents

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

  • โ€ข14 famous open source orgs
  • โ€ขOne wire per day
  • โ€ขPublic, generic
  • โ€ขRead on the web, when you remember

Your wire

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

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