-
Notifications
You must be signed in to change notification settings - Fork 34.4k
Pull requests: huggingface/transformers
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix(zamba): add use_associative_scan config flag to avoid torch.compile slowdown
#48331
opened Aug 26, 2026 by
msnliu
Loading…
Narrow Sentencepiece deserialization exception handling
#48326
opened Aug 26, 2026 by
samyuktahegde
Contributor
Loading…
2 of 6 tasks
Fix Zamba2 construction for num_mem_blocks > 1 checkpoints
#48325
opened Aug 26, 2026 by
john-rocky
Loading…
[Umbrella] Long-context SFT fixes (do not merge)
#48324
opened Aug 26, 2026 by
qgallouedec
Member
•
Draft
[serge] Fix 12 integration tests for model
edgetam failing with import_or_config (other (12))
#48322
opened Aug 25, 2026 by
sergereview
Bot
Loading…
Ignore a stale best checkpoint recorded in a resumed trainer state
#48319
opened Aug 25, 2026 by
VaggelisGian
Contributor
Loading…
4 of 5 tasks
fix(cli): migrate chat.py from requests to httpx to fix transformers env
#48318
opened Aug 25, 2026 by
sundeep8967
Loading…
Byte-level tokenizer.json + Hub tokenizer_class LlamaTokenizer/Fast
#48317
opened Aug 25, 2026 by
itazap
Collaborator
Loading…
fix(mluke): resolve ent ids via vocab not positional index
#48316
opened Aug 25, 2026 by
shoemoney
Loading…
Quiet continuous batching at default verbosity
#48314
opened Aug 25, 2026 by
qgallouedec
Member
Loading…
Fix Moshi output fields
#48311
opened Aug 25, 2026 by
ryota-komatsu
Contributor
Loading…
3 of 6 tasks
Wait for the first request in the async continuous batching bootstrap
#48304
opened Aug 25, 2026 by
qgallouedec
Member
Loading…
Create the continuous batching CPU group with local synchronization
#48302
opened Aug 25, 2026 by
qgallouedec
Member
Loading…
Fail fast when the continuous batching background thread has died
#48301
opened Aug 25, 2026 by
qgallouedec
Member
Loading…
Restore BC for the tensor-parallel API
#48300
opened Aug 25, 2026 by
ArthurZucker
Collaborator
Loading…
Resolve continuous batching config against the text config for composite models
#48299
opened Aug 25, 2026 by
qgallouedec
Member
Loading…
Fall back to the base attention implementation when a paged attention forward has no cache
#48297
opened Aug 25, 2026 by
qgallouedec
Member
Loading…
[Umbrella] Continuous batching fixes for training-while-generating on one weight copy
#48294
opened Aug 25, 2026 by
qgallouedec
Member
•
Draft
Assert cached decode matches recomputing without a cache
#48289
opened Aug 25, 2026 by
IlyasMoutawwakil
Member
Loading…
6 tasks
Infer config class default fields without init a dummy instance
#48282
opened Aug 25, 2026 by
zucchini-nlp
Member
Loading…
Fix DFlash sampled candidates losing the batch dimension
#48281
opened Aug 25, 2026 by
VaggelisGian
Contributor
Loading…
2 of 5 tasks
Fix over-wide KV states from recorded sliding window cache updates
#48280
opened Aug 25, 2026 by
VaggelisGian
Contributor
Loading…
2 of 5 tasks
Previous Next
ProTip!
Filter pull requests by the default branch with base:main.