huggingface/transformers

Pull Requests Commits

vasqu committed 58 days ago

26a7d944

Fix yarn `mscale_all_dim` for DeepSeek v2 and Mistral 4 (#47435)

hmellor committed 58 days ago

Verified 4a1c4b1a

Fix recurrent-layer padding mask being skipped on continued forwards (chunked prefill, cache continuation) (#47087)

abcgco committed 58 days ago

Verified 8c22ce0b

Hoist special-token lookups out of per-token loops in six slow tokenizers (#47425)

ishan-1010 committed 58 days ago

Verified 71554973

fix typos and variable naming in quicktour.md (#47418)

yashasvi-srivastava21 committed 58 days ago

Verified 26e2c764

Normalize multimodal input keys in AnyToAnyPipeline (#47074)

Sunt-ing committed 58 days ago

Verified 9a0fe3f5

Fix Aria checkpoint key conversion mapping (#47151)

sywangyi committed 58 days ago

Verified 16679796

extend tests/models/qwen3_next/test_modeling_qwen3_next.py::Qwen3Next… (#47184)

sywangyi committed 58 days ago

Verified 0248dfda

CI: add serge verify (GPU) caller workflow (#47381)

tarekziade committed 61 days ago

Verified 7ea2320c

ci: harden daily CI reporting against GitHub API rate limits (#47382)

tarekziade committed 61 days ago

Verified c7a96574

Fix `BlockMask` crash in CTRL flex-attention generation (#46854)

jiqing-feng committed 61 days ago

Verified 521f76a3

Fix orthogonal_ init for low-precision dtypes (bf16/fp16) (#47252)

janbernloehr committed 61 days ago

Verified 290d5c4c

[docs] Fix decode examples and expected output in fast_tokenizers (#47369)

samyuktahegde committed 61 days ago

Verified a20c75ce

Fix image-text-to-text stop_sequence handling (#47032)

Sunt-ing committed 61 days ago

Verified 13337f29

[serge] Fix 20 integration tests for model `whisper` failing with `output_mismatch` (list output differs (10), other (6) (#47150)

sergereview[bot] committed 61 days ago

Verified 75a5542e

[Mistral] Move MistralConverter into integrations/mistral/ package (#46603)

juliendenize committed 61 days ago

Verified ab78dc9f

Fix GLM video frame padding for temporal patches (#47141)

labAxiaoming committed 61 days ago

Verified c2ab4167

Refactor image loading in tests to use load_test_image helper (#47218)

LevelVoid committed 61 days ago

Verified 760d519a

Add Cosmos3 Edge model support (#47181)

atharvajoshi10 committed 62 days ago

Verified 150eb7c9

[docs] Inkling (#47350)

stevhliu committed 62 days ago

Verified d9cb3939

Fix Daily CI reporting issues (#47364)

tarekziade committed 62 days ago

Verified 191c44a7

[AudioFlamingo3] Fix bfloat16 dtype mismatch in audio encoder positional embedding (#47258)

snkii committed 62 days ago

Verified b9320048

[`peft`] Support key_mapping with PEFT models (#46766)

tomaarsen committed 62 days ago

Verified 0f33294a

Fix model tests for tipsv2 (#47356)

kaixuanliu committed 62 days ago

Verified 7120f633

Fix TimesFM 2.5 window_size AttributeError (#47363)

kashif committed 62 days ago

Verified 5fbd5f03

fix: allow num_labels property to return None when id2label is unset (#47069)

SebTardif committed 62 days ago

Verified d9f38e80

[Tests] Fix slow video tensor creation from list of numpy arrays in SmolVLM (#44731)

Defalt-Meh committed 62 days ago

Verified fe6ff6c4

Update dev version on main (#47366)

vasqu committed 62 days ago

Verified a8609bed

[CB] Automatically switch attention implementation to flash (#47330)

remi-or committed 62 days ago

Verified 279da5ce

Fix deepgemm on multiple devices (#47323)

IlyasMoutawwakil committed 62 days ago

Verified 01b03b2e

Older