-
Notifications
You must be signed in to change notification settings - Fork 33.3k
Pull requests: huggingface/transformers
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix model parallel device mismatch issue in
create_bidirectional_mask
#46221
opened May 26, 2026 by
kaixuanliu
Contributor
Loading…
[
Configs] Fix layer type validation to include its mlp counterpart
#46220
opened May 26, 2026 by
vasqu
Contributor
Loading…
fix(hrm_text): Add XPU Expectations for tests
#46214
opened May 26, 2026 by
kaixuanliu
Contributor
•
Draft
Bump the actions group with 19 updates
dependencies
Pull requests that update a dependency file
github_actions
Pull requests that update GitHub Actions code
#46212
opened May 26, 2026 by
dependabot
Bot
Loading…
fix(llama): allow explicit head_dim when hidden_size not divisible by num_attention_heads
#46211
opened May 26, 2026 by
Sriniketh24
Contributor
Loading…
1 task done
Fix bnb 4bit/8bit quantization drop chunked tensors bug
#46210
opened May 26, 2026 by
kaixuanliu
Contributor
Loading…
Replace source-reading in _can_set_*_implementation with class introspection
#46207
opened May 26, 2026 by
rasmi
Contributor
Loading…
3 of 6 tasks
Guard DeviceMesh import in continuous batching
#46205
opened May 26, 2026 by
danyalahmed1995
Loading…
Fix
num_items_in_batch over-counting for causal LM losses
#46204
opened May 26, 2026 by
qgallouedec
Member
Loading…
[deepseek_v4] keep hc_head / sinks / position_bias in fp32
#46198
opened May 25, 2026 by
ArthurZucker
Collaborator
Loading…
[SeamlessM4T/v2] Support attn_implementation=sdpa dispatch
#46196
opened May 25, 2026 by
YangKai0616
Contributor
Loading…
Fall back to flat kwarg when modality dict is passed without it
#46195
opened May 25, 2026 by
Ace3Z
Loading…
5 tasks done
Tighten exception handling in find_sentencepiece_model_file
#46177
opened May 24, 2026 by
Aditya-ad48
Loading…
2 of 5 tasks
[Gemma4] Replace one-hot matmul with F.embedding in position embeddings
#46176
opened May 24, 2026 by
Sriniketh24
Contributor
Loading…
fix(mps): build_2d_sinusoidal_position_embedding crashes on Apple Silicon due to float64 on MPS device
#46174
opened May 23, 2026 by
shubhammr21
Loading…
Fix StaticLayer.get_seq_length return type annotation (#45987)
#46173
opened May 23, 2026 by
Sanjays2402
Loading…
3 tasks done
[LED][Longformer] Replace for-loop with unfold in _chunk ONNX-export path
#46169
opened May 23, 2026 by
guinik
Loading…
3 of 6 tasks
[OpenAI Privacy Filter] banded SWA eager attention (O(N*W) instead of O(N^2))
#46168
opened May 23, 2026 by
kiankyars
Loading…
Romanian translation of README.md, index.md, installation.md, _config.py and quicktour.md.
#46166
opened May 22, 2026 by
filipinescu
Loading…
2 tasks done
add XPU Expectations for florence2 and lfm2_vl model test
#46162
opened May 22, 2026 by
kaixuanliu
Contributor
•
Draft
Previous Next
ProTip!
Add no:assignee to see everything that’s not assigned.