huggingface/transformers

Pull Requests Commits

Trigger CI: full important models list for pr49425

ydshieh committed 23 days ago

a49ab794

Trigger CI: pr49425 on 536ecc007, vit-only dummy run

ydshieh committed 23 days ago

121029dd

review(test): relax whisper expectations on TPU

tengomucho committed 23 days ago

9a476110

test: add TPU expectations to the whisper integration tests

tengomucho committed 23 days ago

ad119219

test: add TPU expectations to the mistral3 integration tests

tengomucho committed 23 days ago

52e892b7

test: add TPU expectations to the internvl integration tests

tengomucho committed 23 days ago

e9981f2a

test: add TPU expectations to the llama integration tests

tengomucho committed 23 days ago

56f98433

test: add TPU expectations to the qwen2 integration tests

tengomucho committed 23 days ago

eb2123f3

test: add TPU expectations to the gpt2 integration tests

tengomucho committed 23 days ago

f930045d

test: add TPU expectations to the gemma3n integration tests

tengomucho committed 23 days ago

5ba2e5fb

test: add TPU expectations to the gemma3 integration tests

tengomucho committed 23 days ago

eb677c74

test: restore the default device when the global device test fails

tengomucho committed 23 days ago

30183c2f

test: run subprocess tests in-process on single-process devices

tengomucho committed 23 days ago

d6bade68

test: request full-precision float32 matmuls in the test suite

tengomucho committed 23 days ago

38bcf73f

test: use the CUDA SDPA tolerances on TPU

tengomucho committed 23 days ago

3d4622a0

Second phase of torch 2.6 cleanups (#49396)

Rocketknight1 committed 25 days ago

Verified 536ecc00

Use the Hub rotary kernel for inference only on CUDA (#49439)

qgallouedec committed 25 days ago

Verified 4f2db330

Guard the checks for current_accelerator() and bump Torch version (#49395)

Rocketknight1 committed 25 days ago

Verified 8138fc45

Use one static Transformers CI badge in PR descriptions (#49471)

tarekziade committed 25 days ago

Verified a7e8b33b

fix: treat every declared field of the config class as a default fixes #49456 (#49466)

tarekziade committed 25 days ago

Verified 9c59c0c6

Keep AutoVideoProcessor class lookup lazy (#49458)

DimensionSTP committed 25 days ago

Verified 927af2f7

[MiMo-V2-Flash] Count the query token in the sliding window (#49170)

guapisolo committed 25 days ago

Verified 429e6916

Add Molmo2 (#43451)

SangbumChoi committed 25 days ago

Verified f27b9b14

Ci fetches multimodal models recursively (#49389)

zucchini-nlp committed 25 days ago

Verified 90ef040d

Recurse for all subconfigs (#49461)

zucchini-nlp committed 25 days ago

Verified 5c9f079c

Aria, Ernie4.5-VL-MoE, Gemma4, Inkling, MiniMax-M3-VL, Qwen3.5-MoE: run the model tests on the VLM/ALM testers (#49404)

IlyasMoutawwakil committed 25 days ago

Verified c48824e8

Image decoding moves to torchcodec (#47963)

zucchini-nlp committed 25 days ago

Verified c0586068

Fix CI slack reporting: surface crashed/missing model jobs in model_results.json (#49427)

ydshieh committed 26 days ago

Verified 65dc402e

[Atlas Inference] Enable torch.compile for the Strix Halo Gated DeltaNet hub kernel (#49453)

AzeezIsh committed 26 days ago

Verified 33ab420a

[Atlas Inference] Fix GB10 Gated DeltaNet hub kernel for transformers 5.17+ and enable torch.compile (#49452)

AzeezIsh committed 26 days ago

Verified 2215242b

Older