huggingface/transformers

Pull Requests Commits

More skips/removals

Rocketknight1 committed 257 days ago

ca8c8df6

More skips/removals

Rocketknight1 committed 257 days ago

b8c560d2

Skip a flaky test

Rocketknight1 committed 257 days ago

d1a1888a

Fixes DynamicCache export issues due to control flow and inplace modifications (#36652)

xadupre committed 257 days ago

Verified 6f5dc9c8

Add device workaround for int4 weight only quantization after API update (#36980)

jerryzh168 committed 257 days ago

Verified a1654589

Skip code `307` in `RequestCounter` (#36953)

ydshieh committed 257 days ago

Verified ed95493c

[chat-template] fix video loading (#37146)

zucchini-nlp committed 257 days ago

Verified 211e4dc9

[doc] Fix link for Quark quantization page (#37179)

BowenBao committed 258 days ago

Verified 800510c6

Revert #37031 (#37178)

Cyrilvallez committed 258 days ago

Verified 41f5c321

Fix meta state dict loading with quantizers (#37136)

Cyrilvallez committed 258 days ago

Verified bc2dea3f

Avoid pipeline test failing related to Hub call (#37170)

ydshieh committed 258 days ago

Verified 35253076

Fixes the inconsistency of the optionality of attention_mask (#37153)

Zephyr271828 committed 258 days ago

Verified bf41e54f

Refactor attention for SigLIP based models (#36981)

qubvel committed 258 days ago

Verified 3249c5dc

fix XPU UT error case brough by RNG difference btw XPU and CUDA (#37121)

yao-matrix committed 258 days ago

Verified 24e311f4

[`ModernBERT`] Never save 'reference_compile' config; should be set based on end user (#36305)

tomaarsen committed 258 days ago

Verified 897ff9af

Make canine model exportable by removing unncessary complicated logic (#37124)

tugsbayasgalan committed 258 days ago

Verified c0bd8048

Only count num items in batch when needed (#36867)

IlyasMoutawwakil committed 258 days ago

Verified 60b75d99

Convert `_VALID_DICT_FIELDS` to class attribute for shared dict parsing in subclasses (#36736)

Tavish9 committed 258 days ago

Verified fac70ff3

Use public export API on torch 2.5 and future (#36781)

guangy10 committed 258 days ago

Verified ae34bd75

enable `test_assisted_decoding_in_different_gpu` test on XPU (#37120)

yao-matrix committed 258 days ago

Verified 8f6b27eb

Fix llava xpu tests. (#37130)

jiqing-feng committed 258 days ago

Verified 737cbd21

add gpt2 test on XPU (#37028)

jiqing-feng committed 258 days ago

Verified 3a6ab46a

Fix std initialization in Idefics variants (#37100)

yaswanth19 committed 258 days ago

Verified 4b13a029

Fix more inefficient PT operations (#37060)

cyyever committed 259 days ago

Verified 786d9c5e

Refactor `return_dict` logic to remove complicated if/else paths (#36794)

qubvel committed 259 days ago

Verified a1e389e6

Remove low_cpu_mem_usage and _fast_init (#36963)

Cyrilvallez committed 259 days ago

Verified f304318f

[qwen3] fix generation tests (#37142)

zucchini-nlp committed 259 days ago

Verified 88056004

[Feature] Support using FlashAttention2 on Ascend NPU (#36696)

FightingZhen committed 259 days ago

Verified e686fed6

ydshieh committed 259 days ago

Verified a03cee7a

Export T5 (encoder-decoder) to ExecuTorch (#36486)

guangy10 committed 259 days ago

Verified 3b07ca78

Older