huggingface/transformers

Pull Requests Commits

Add analyze_mem.py and memory_tracker_plugin from debug_mem

ydshieh committed 1 hour ago

ffee7dfc

ydshieh committed 16 hours ago

ecafb707

tests/deepseek_ocr2,got_ocr2: add tiny_model_id, remove _setup_tokenizer

ydshieh committed 17 hours ago

cff4aa82

tests/llama4: add tiny_model_id, remove _setup_tokenizer hook

ydshieh committed 17 hours ago

7a05ecf2

tests/kimi_k25: add tiny_model_id to speed up processor tests

ydshieh committed 18 hours ago

b6c30e53

Reject path traversal in the `transformers_weights` config field (#46890)

LinZiyuu committed 19 hours ago

Verified bce8fd08

[docs] Custom code conversion mapping (#47114)

stevhliu committed 20 hours ago

Verified 3bbac3e0

Add exporters min version requirements and test skip (#47161)

IlyasMoutawwakil committed 20 hours ago

Verified d15c0361

tests: reduce processor test memory usage and use tiny test assets (#47168)

ydshieh committed 23 hours ago

Verified 41fc5b47

Clarify input device placement in the Quicktour inference example (#47136)

samyuktahegde committed 23 hours ago

Verified b7653886

Extend continuous batching memory prediction test to XPU (#47159)

sywangyi committed 1 day ago

Verified e8cfae1d

Fix case where `_LazyAutoMapping.register` is passed a `str` key (#47148)

hmellor committed 1 day ago

Verified 73cde7c0

[docs] MoE decode switching (#47149)

stevhliu committed 1 day ago

Verified 772f5400

add XPU output expectations for minicpm3 tests (#47092)

kaixuanliu committed 1 day ago

Verified 6f5f5b4d

Diffusion gemma: fix failed test cases (#47025)

kaixuanliu committed 1 day ago

Verified 7a69c513

add XPU Expectation for cosmos3_omni tests (#46880)

kaixuanliu committed 1 day ago

Verified 8e502a17

Fix IndexError Bug in XLMRoberta/Camembert ForMultipleChoice by restoring the pooler (#47147)

pariidanDKE committed 1 day ago

Verified 187fda1b

Restrict QuantizedCache to only full attention (#47160)

Cyrilvallez committed 1 day ago

Verified d876a49f

Skip caching_allocator_warmup on Neuron (no reuse pool to warm; currently OOMs) (#47029)

dacorvo committed 1 day ago

Verified d49affb3

[docs] continuous batching (offloading behavior, max batch tokens, block size minimum) (#46925)

stevhliu committed 1 day ago

Verified 4af901d5

Fix MoE decode optimization for experts living in a submodel (#47107)

IlyasMoutawwakil committed 1 day ago

Verified 1d828044

[docs] fix autolinks (#46968)

stevhliu committed 1 day ago

Verified a92059d9

Adapt voxtral_realtime to mistral-common 1.11.5 audio API (#46920)

juliendenize committed 1 day ago

Verified 8e324030

revert #47121 (#47144)

eustlb committed 1 day ago

Verified 67538ec6

Fix output labels for AudioFlamingo3 (and related) models (#47112)

ebezzam committed 1 day ago

Verified 080fe1e0

Fix false __len__ claims in Trainer docstrings (#47131)

albertvillanova committed 1 day ago

Verified 0b38752f

processor tests: use tiny Hub repos to reduce CI memory (#47115)

ydshieh committed 2 days ago

Verified 001c526a

Fix: accept scalar audio formats in Voxtral transcription requests (#47045)

Sunt-ing committed 2 days ago

Verified b406cede

[serge] Fix 12 integration tests for model `dac` failing with `output_mismatch` (tensor values differ (6), other (6)) (#47121)

sergereview[bot] committed 2 days ago

Verified c2e10144

Fix CLI compatibility with huggingface_hub 1.22 (#47059) (#47064)

dhruv7477 committed 2 days ago

Verified 326ae42d

Older