Skip to content

Pull requests: huggingface/transformers

New pull request New
540 Open 18,038 Closed
540 Open 18,038 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Fix Qwen2RotaryEmbedding Device Matching Bug
#35506 opened Jan 4, 2025 by developer0hye Loading…
2 of 5 tasks
1
Add autoquant support for torchao quantizer
#35503 opened Jan 3, 2025 by jerryzh168 Loading…
2
Split and clean up GGUF quantization tests
#35502 opened Jan 3, 2025 by Isotr0py Loading…
1 of 5 tasks
7
Update conftest.py
#35500 opened Jan 3, 2025 by tauhee56 Loading…
5 tasks done
2
Replace deprecated batch_size with max_batch_size when using HybridCache
#35498 opened Jan 3, 2025 by mtreinik Loading…
1 of 5 tasks
1
run_audio_classification.py freeze_base_model in w2v-bert
#35497 opened Jan 3, 2025 by HERIUN Loading…
3 of 5 tasks
1
🌐 [i18n-KO] Remove duplicates in toctree
#35496 opened Jan 3, 2025 by jungnerd Loading…
2 of 5 tasks
Optimized set_initialized_submodules.
#35493 opened Jan 3, 2025 by LagPixelLOL Loading…
Update torchao.md: use auto-compilation
#35490 opened Jan 2, 2025 by martin0258 Loading…
4 of 5 tasks
1
5
[tokenizer] fix llama tokenizer (slow)
#35488 opened Jan 2, 2025 by DrownFish19 Loading…
3 of 5 tasks
Make warmup_min_lr configurable in DeepSpeed integration
#35487 opened Jan 2, 2025 by Tanuj-rai Loading…
5 tasks
1
3
use warning_once instead of warning in Trainer.tokenizer
#35482 opened Jan 2, 2025 by jzhang533 Loading…
2 of 5 tasks
1
Update llm_optims docs for sdpa_kernel
#35481 opened Jan 2, 2025 by jla524 Loading…
Bump torch requirement to 2
#35479 opened Jan 2, 2025 by oraluben Loading…
1 of 5 tasks
Add helper for torch dynamo
#35478 opened Jan 2, 2025 by oraluben Loading…
1 of 5 tasks
1
Add FAST
#35476 opened Jan 1, 2025 by jadechoghari • Draft
cover edge case where decoder_input_ids is provided but it is empty
#35475 opened Jan 1, 2025 by adi Loading…
Add SDPA support for LayoutLMv3 model
#35469 opened Dec 31, 2024 by stancld Loading…
3 of 4 tasks
fix: Qwen2-VL generate with inputs_embeds
#35466 opened Dec 31, 2024 by minostauros Loading…
2 of 5 tasks
1
1
ModernBert: reuse GemmaRotaryEmbedding via modular + Integration tests
#35459 opened Dec 30, 2024 by tomaarsen Loading…
2 of 5 tasks
2
fix document qa bf16 pipeline
#35456 opened Dec 30, 2024 by jiqing-feng Loading…
2
Fix #35447 Tokenizer does not split text according to newly added input tokens
#35455 opened Dec 29, 2024 by jiongjiongli Loading…
Support constant lr with cooldown
#35453 opened Dec 29, 2024 by LoserCheems Loading…
3 of 5 tasks
1
4
Replace tokenizer to processing_class in Seq2SeqTrainer
#35452 opened Dec 29, 2024 by zzaebok Loading…
2 of 5 tasks
2
T5 static cache
#35445 opened Dec 29, 2024 by cptspacemanspiff Loading…
ProTip! Follow long discussions with comments:>50.