Pull Requests huggingface/transformers

docs: wrap sam3-litetext inference example in torch.inference_mode()

#46391 opened 2026-06-03 21:41 by kpal002

Fix typo in mm_grounding_dino converter

#46390 opened 2026-06-03 21:30 by professoralbay

HF Buckets integration to Trainer

#46386 opened 2026-06-03 16:36 by SunMarc

DSv4 static cache

#46384 opened 2026-06-03 14:14 by IlyasMoutawwakil

Add vLLM smoke test to CI

#46383 opened 2026-06-03 13:04 by hmellor

Allow tokenizers 0.23.1

#46381 opened 2026-06-03 12:15 by ArthurZucker

Fix cis

#46380 opened 2026-06-03 12:07 by ArthurZucker

Update hub metadata

#46379 opened 2026-06-03 11:54 by zucchini-nlp

[CB] [Major] Add multimodal support to continous batching

#46376 opened 2026-06-03 10:27 by remi-or

[VITS] Un-skip test_model_forward

#46375 opened 2026-06-03 09:56 by blipbyte

Better processing tests

#46374 opened 2026-06-03 09:19 by zucchini-nlp

Fix: interleaved RoPE application for MLA and Support Index Cache DSA indexer skip-topk sharing for GLM5

#46372 opened 2026-06-03 08:39 by Dovis01

skip invalid quant_cache test for nemotron_h

#46368 opened 2026-06-03 07:23 by kaixuanliu

extend deepseek v4 test to xpu

#46366 opened 2026-06-03 05:28 by sywangyi

chore(linter): update DataClass type definitions to TypeAlias

#46364 opened 2026-06-03 05:18 by cheparity

skip quant cache tests for deepseek v4

#46362 opened 2026-06-03 02:39 by kaixuanliu

[docs] deepgemm

#46361 opened 2026-06-03 00:57 by stevhliu

[docs] gguf mps

#46357 opened 2026-06-02 21:43 by stevhliu

Test models instantiate from default config on meta + fix failing models

#46355 opened 2026-06-02 19:12 by ArthurZucker

[docs] metal flash attention

#46349 opened 2026-06-02 17:25 by stevhliu

Add TIPSv2

#46347 opened 2026-06-02 14:35 by Ternura143

fix(falcon_mamba): recommend `pip install kernels` in fast-path warning

#46343 opened 2026-06-02 13:11 by Anai-Guo

Fix tokenizer keep_accents defaults and update integration tests

#46341 opened 2026-06-02 12:33 by itazap

Extended & simplified n-to-1 kernel fusion via KernelConfig

#46339 opened 2026-06-02 09:21 by michaelbenayoun

Fix tests: 'Cohere2MoeModel' object has no attribute 'hf_device_map'

#46337 opened 2026-06-02 07:32 by kaixuanliu

Add Nemotron ASR Streaming

#46332 opened 2026-06-01 17:04 by eustlb

Add Parakeet-RNNT

#46331 opened 2026-06-01 16:57 by eustlb

Fix: strip EOS token when continue_final_message=True in MistralCommo…

#46330 opened 2026-06-01 16:29 by ranjithchinnanan

feat(quantizers): support non-weight param names in TorchAo safetensors loading

#46325 opened 2026-06-01 12:58 by agesf

Quantize models on `__init__` / `from_config`

#46324 opened 2026-06-01 12:33 by ArthurZucker