huggingface/transformers

Pull Requests Commits

Support HSDP with expert parallelism

3outeille committed 3 hours ago

015a1fb8

Merge branch 'main' into hsdp

3outeille committed 4 hours ago

Verified c10d56be

Chat template output tensor type (#50103)

zucchini-nlp committed 6 hours ago

Verified 49368678

Add HSDP (dp_replicate_size) to transformers.distributed

3outeille committed 7 hours ago

9fb97b9d

[Distributed] Set `device_id` only for NCCL (#50281)

michaelbenayoun committed 7 hours ago

Verified 409c3490

[Trainer] Compile with static shapes when the backend is neuron (#49617)

qgallouedec committed 11 hours ago

Verified 26754696

[Improvement] Formalize KV sharing (#47290)

remi-or committed 14 hours ago

Verified cc1451df

Docs: remove duplicate hf_token secret in build_documentation.yml (#50168)

mishig25 committed 15 hours ago

Verified 4888db4e

Fix `bboxes` being overwritten before use in `prepare_example` (#50218)

Jose-Marcial-GF committed 16 hours ago

Verified c97587a4

fix(tests_fetcher): handle `X as X` re-export syntax and duplicate module entries (#50253)

ydshieh committed 17 hours ago

Verified e03c3539

Read add_bos_token and add_eos_token from GGUF files (#49649)

francip committed 18 hours ago

Verified 7c22a03f

[serve] Respect max_completion_tokens in /v1/chat/completions (#50225)

fruitstash committed 18 hours ago

Verified a9124a87

Dequantize MXFP4 experts to the requested dtype (#50256)

qgallouedec committed 18 hours ago

Verified 67685971

deal with Nan values for aggregating loss (#50172)

SunMarc committed 18 hours ago

Verified b47ec788

Fix OOM regression for FSDP2 + cpu_ram_efficient_loading on large models (#45649)

AmineDiro committed 19 hours ago

Verified 5949d2c4

[serge] Fix 2 integration tests regressed by commit 7d3ae42ee810 (PR #49811) (#50098)

sergereview[bot] committed 20 hours ago

Verified 763b4ba5

Fix resize_embeddings tests for deepspeed (#49909)

kaixuanliu committed 20 hours ago

Verified 28aa8646

[VibeVoice] Add `sample` argument to disable VAE noise in audio features (#50179)

hmellor committed 21 hours ago

Verified 2f2e711f

fix failed test cases for pocket_tts model (#50159)

kaixuanliu committed 22 hours ago

Verified 3e8d3971

fix(tests_fetcher): use posix paths throughout, eliminate all os.path.sep/os.path.join patterns (#50219)

ydshieh committed 22 hours ago

Verified a8f6bda5

Remove the GitHub API cache-bust from the AMD docker images (#49351)

tarekziade committed 22 hours ago

Verified 044aa4c8

[Gemma4] Pass num_frames_per_video to get_video_features in forward (#50237)

yzkkai committed 1 day ago

Verified 5f6629d8

fix Load DETR-family panoptic masks with load_image_as_tensor (#50234)

dhekimian committed 1 day ago

Verified 9167cfce

update to torch 2.15 for xpu (#50236)

sywangyi committed 1 day ago

Verified 7db0255f

Use repo-wide linguist-generated globs so new models don't touch .gitattributes (#50181)

ArthurZucker committed 1 day ago

Verified 98110a1f

Fix/trainer parallel token count 50212 (#50214)

BlazeSR10hawk committed 1 day ago

Verified 9f8b2169

Modular suffix overlap - allow non-model classes (#49754)

zucchini-nlp committed 1 day ago

Verified e2e6cc93

[PaliGemma] Apply final logit softcapping (#50156)

alejandrorodriguezalvarez884-dot committed 1 day ago

Verified 3a667d7c

Audio from video in jinja (#49382)

zucchini-nlp committed 1 day ago

Verified e400105d

[CI] Docs bot (#49999)

stevhliu committed 2 days ago

Verified ca8c871e

Older