-
Notifications
You must be signed in to change notification settings - Fork 33.6k
Pull requests: huggingface/transformers
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
docs(conditional_detr): fix num_queries default in docstring (100 -> 300)
#46939
opened Jun 27, 2026 by
Kropiunig
Loading…
Add unit tests for
_ignore_causal_mask_sdpa in test_masking_utils.py
#46938
opened Jun 27, 2026 by
MosadCreates
Loading…
Reject assisted generation for LFM2 and LFM2-MoE (set _is_stateful)
#46937
opened Jun 27, 2026 by
Sunt-ing
Contributor
Loading…
3 of 6 tasks
Fix crash in greedy assisted generation with different tokenizers
#46936
opened Jun 27, 2026 by
Sunt-ing
Contributor
Loading…
1 task done
Fix
TrackioCallback fails to log evaluation metrics after training ends
#46935
opened Jun 27, 2026 by
lewtun
Member
Loading…
2 of 6 tasks
Add attention mask skip tests, tokenizer padding edge case tests, and…
#46934
opened Jun 27, 2026 by
MosadCreates
Loading…
6 tasks
[Gemma4] Fix dtype casting for quantized vision/audio embedders
#46933
opened Jun 27, 2026 by
sharmax-vikas
Contributor
Loading…
1 of 3 tasks
Bump minimum torch version to 2.5 and clean up torch<2.5 workarounds
#46930
opened Jun 27, 2026 by
cyyever
Contributor
Loading…
1 of 6 tasks
Fix torch.export compatibility for Mixtral MoE experts
#46929
opened Jun 27, 2026 by
Sarimsaljook
Loading…
3 of 6 tasks
[serge] Fix 8 integration tests for model
blip_2 failing with other (other (8))
#46928
opened Jun 27, 2026 by
sergereview
Bot
Loading…
[serge] Fix 12 integration tests for model
glm46v failing with other (other (12))
#46927
opened Jun 27, 2026 by
sergereview
Bot
Loading…
[serge] Fix 12 integration tests for model
minimax_m3_vl failing with load_error (other (12))
#46926
opened Jun 27, 2026 by
sergereview
Bot
Loading…
[docs] continuous batching (offloading behavior, max batch tokens, block size minimum)
#46925
opened Jun 26, 2026 by
stevhliu
Member
Loading…
Replace VideoMAE sinusoid encoding helper with PyTorch implementation
#46924
opened Jun 26, 2026 by
praful-srinivasan-027
Contributor
•
Draft
docs: add prepare_for_model() v4→v5 migration example
#46921
opened Jun 26, 2026 by
MushiSenpai
Loading…
6 tasks
Adapt voxtral_realtime to mistral-common 1.11.5 audio API
#46920
opened Jun 26, 2026 by
juliendenize
Contributor
Loading…
1 of 6 tasks
Fix RagTokenizer attribute delegation
#46919
opened Jun 26, 2026 by
Pruthvi226
Loading…
5 of 6 tasks
fix(deepspeed): chunk ZeRO-3 missing-key param gather to avoid OOM
#46918
opened Jun 26, 2026 by
itxsamad1
Loading…
1 of 2 tasks
Remove deprecated training args
#46917
opened Jun 26, 2026 by
cyyever
Contributor
Loading…
1 of 6 tasks
Fix training-loss double-shift in Florence2 and CohereASR
#46916
opened Jun 26, 2026 by
muhamedfazalps
Loading…
Validate shard filenames in checkpoint index to prevent path traversal (silent weight injection)
#46913
opened Jun 26, 2026 by
Snakinya
Loading…
[serge] Fix 8 integration tests for model
deepseek_v32 failing with load_error (other (8))
#46908
opened Jun 26, 2026 by
sergereview
Bot
Loading…
Previous Next
ProTip!
no:milestone will show everything without a milestone.