Skip to content

Issues: huggingface/transformers

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

1,006 Open 15,561 Closed
1,006 Open 15,561 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
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

FSDP Torch XLA vs. FSDPv2 (SMPD) Torch XLA checkpoint saving bug bug
#36004 opened Feb 1, 2025 by salrowili
4 tasks
Mismatch Between Image Tokens and Features in LLaVA Model Fine-Tuning
#36002 opened Feb 1, 2025 by Md-Nasif03
past_key_values not being set in model_inputs keys bug
#36001 opened Feb 1, 2025 by AEJaspan
1 of 4 tasks
Feature Request: add disable span CLI flag Feature request Request for a new feature
#35999 opened Feb 1, 2025 by obloomfield
Unable to load openai/whisper-tiny bug
#35995 opened Jan 31, 2025 by pprobst
2 of 4 tasks
model.parameters() return [Parameter containing: tensor([], device='cuda:0', dtype=torch.bfloat16, requires_grad=True)] when using zero3 bug
#35994 opened Jan 31, 2025 by fanfanffff1
2 of 4 tasks
Transformers PaliGemma evaluate and compute_loss fail with tensors/device errors bug Cache
#35990 opened Jan 31, 2025 by BlGene
4 tasks
7
Add support for context parallelism Feature request Request for a new feature
#35983 opened Jan 31, 2025 by lewtun
@ydshieh
1
Docs: return type of get_default_model_and_revision might be incorrectly documented?
#35981 opened Jan 31, 2025 by MarcoGorelli
1
1
HPD-Transformer: A Hybrid Parsing-Density Transformer for Efficient Structured & Probabilistic Reasoning New model
#35978 opened Jan 31, 2025 by infodevlovable
2 tasks done
adalomo and deepspeed zero3 offload error bug
#35977 opened Jan 31, 2025 by YooSungHyun
2 of 4 tasks
4
Deformable DETR custom kernel fails to compile with PyTorch 2.6 bug Vision
#35976 opened Jan 30, 2025 by hassonofer
4 tasks
1
2
[Phi-3-mini-128k-instruct] Difference of encodings for Slow and Fast Tokenizer bug
#35973 opened Jan 30, 2025 by GKIBMNY
4 tasks
1
Cannot import 'GenerationOutput' in 4.48.1 bug
#35957 opened Jan 29, 2025 by inthree3
2 of 4 tasks
2
Can't use Trainer on mps device bug
#35954 opened Jan 29, 2025 by smartliuhw
2 of 4 tasks
1
Mixed-precision with torch.autocast is broken for many models when using attn_implementation="flash_attention_2" bug
#35945 opened Jan 28, 2025 by konstantinjdobler
2
Learning rate logging off by one training step bug
#35942 opened Jan 28, 2025 by phoerious
2 of 4 tasks
1
Mangled tokenization with Llama 3.1 for string sequences containing<space>'m bug
#35938 opened Jan 28, 2025 by tomjorquera
2 of 4 tasks
3
Llama3.2: Allow batch to have Feature request Request for a new feature VLM
#35937 opened Jan 28, 2025 by maximilianmordig
1
Add YuE audio model Good Difficult Issue New model
#35929 opened Jan 28, 2025 by ArthurZucker
2 tasks done
3
Add Deepseek AI's Janus model Good Difficult Issue New model
#35928 opened Jan 28, 2025 by ArthurZucker
2 tasks done
3
4.48.1 breaks sliding window in eager attention for qwen2 bug
#35924 opened Jan 28, 2025 by wizyoung
4 tasks
1
Llama tokenizer newline character inconsistency
#35923 opened Jan 28, 2025 by ingo-m
1
meta-llama/Llama-3.2-11B-Vision-Instruct, device_map = 'auto', padding ruins _prepare_4d_causal_attention_mask_with_cache_position bug
#35918 opened Jan 27, 2025 by AndriiZelenko
2 of 4 tasks
1
1
Attention can be None in ModernBertForSequenceClassification
#35917 opened Jan 27, 2025 by ashmikuz
1
5
ProTip! Follow long discussions with comments:>50.