huggingface/transformers

Pull Requests Commits

add other models

itazap committed 259 days ago

38a8ce45

itazap committed 259 days ago

0a246a5a

itazap committed 259 days ago

0d72f204

rm func from some models

itazap committed 259 days ago

1cb4b7fb

rm build_input function for fast, don't test prepare_for_model for fast as we don't use it

itazap committed 259 days ago

f3fb6164

rm build_input.. from old file

itazap committed 259 days ago

2b8774a7

rm build_inputs_with_special_tokens from llama and gemma

itazap committed 259 days ago

adeb8cdd

itazap committed 259 days ago

148e3159

itazap committed 259 days ago

cc76a4f1

itazap committed 259 days ago

0202f862

[MODEL] Add Falcon H1 (#38249)

younesbelkada committed 259 days ago

Verified 6829936e

tp plan should not be NONE (#38255)

ArthurZucker committed 259 days ago

Verified e288ee00

Revert parallelism temporarily (#38240)

LysandreJik committed 260 days ago

Verified 711d78d1

CI reporting improvements (#38230)

ydshieh committed 260 days ago

Verified feec294d

Protect ParallelInterface

LysandreJik committed 260 days ago

cb513e35

LysandreJik committed 260 days ago

f4ef41c4

[gemma3] fix bidirectional attention mask (#38080)

zucchini-nlp committed 260 days ago

Verified f834d368

[mllama] fix loading and inference (#38223)

zucchini-nlp committed 260 days ago

Verified 2edb0e4b

Add padding-free to bamba (#35861)

garrett361 committed 260 days ago

Verified 390f1534

Fixing Bitnet after use_rms_norm introduction (#38229)

MekkCyber committed 260 days ago

Verified 2a794713

Enable Quantize KV Cache for Mistral Model (#35042)

Bojun-Feng committed 260 days ago

Verified 96618960

parallelism goes brrr (#37877)

NouamaneTazi committed 260 days ago

Verified 1c2f36b4

Fix Llama4 (#38222)

Cyrilvallez committed 260 days ago

Verified b591d925

Mamba2 remove unecessary test parameterization (#38227)

ivarflakstad committed 260 days ago

Verified 3f0b7d0f

Minor llama4 fixes (#38123)

molbap committed 260 days ago

Verified 9cde2f5d

fix dead flax links modeling_flax_pytorch_utils.py (#38212)

dizer-ti committed 260 days ago

Verified 856f034f

Make `train_dataset` attribute in `_get_train_sampler` optional (#38226)

SunMarc committed 260 days ago

Verified bb3c6426

In Llama4 fix wrongly inverted causal attention mask when using SDPA implementation (#38094)

sogartar committed 260 days ago

Verified 2ad152f8

Disable torchscript tests for AriaForConditionalGenerationModelTest (#38225)

ivarflakstad committed 260 days ago

Verified de70c842

Add support to Marimo Notebooks and Enverge.ai (#38210)

brenoca committed 260 days ago

Verified 8ea61c45

Older