Skip to content
Problems.

Hugging Face users keep filing bugs

MLOps15 posts from 15 people13d active+33 posts in the last 7 days, 6 the 7 days before (fading)

Posts per day

Posts per day15 posts, Aug 28 to Sep 25

The posts behind it

15, newest first
PostDate
All paid GPU pools fail to start: MOUNT_ATTR_IDMAP error at container init (Exit code 128)Hugging Face ForumsmiguralexSep 25today
Unable to re-submit the authorship claimHugging Face ForumsHaoboZhengSep 232 days ago
[i18n-<languageCode>] Translating docs to <languageName>Hi! Let's bring the documentation to all the -speaking community 🌐 (currently 0 out of 267 complete) Who would want to translate? Please follow the 🤗 [TRANSLATING guide](https://github.com/huggingface/transformers/blob/main/docs/TRANSLATING.md). Here is a list of the files...huggingface/transformersHALIMMOUSAASep 205 days ago
T5 sentencepiece tokenizer without precompiled_charsmap cannot be loaded### System Info - `transformers` version: 5.18.0.dev0 - Platform: Linux-7.1.10-arch1-1-x86_64-with-glibc2.43 - Python version: 3.14.4 - Huggingface_hub version: 1.31.0 - Safetensors version: 0.8.0 - Accelerate version: 1.15.0 - Accelerate config: not found - DeepSpeed version:...huggingface/transformersDRKV333Sep 187 days ago
gemma 4 can not load audio file and produce Audio features and audio tokens do not match, tokens: 31, features: 468480### System Info I try in kaggle T4 x 2 ### Who can help? _No response_ ### Information - [ ] The official example scripts - [x] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [x] My own task or dataset...huggingface/transformerschaowenguoSep 178 days ago
Issues with multiple spaces stuck with "No @spaces.GPU function detected during startup"Hugging Face ForumsaidemosbapSep 169 days ago
Nothing destroys the process group, so every distributed Trainer run warns at exit### System Info transformers 5.18.0.dev0, accelerate 1.15.0, torch 2.13.0+cu129. ### Who can help? @SunMarc ### Reproduction Any distributed `Trainer` script that exits after training prints, once per rank: The Trainer sets distributed training up through accelerate and nothing...huggingface/transformersqgallouedecSep 169 days ago
Can't join discord and the course links are brokenHugging Face Forumsdominik4workstuffSep 1411 days ago
How do i make and add my api key connected to somethingHugging Face ForumssolaraaaSep 1213 days ago
[zai-org/GLM-OCR] <|image|> token has special=False in Rust tokenizer backend — should it be True?## Problem The \` \` token (ID 59280) in \`zai-org/GLM-OCR\` is listed in \`tokenizer.all_special_tokens\` on the Python side, but its entry in the Rust (\`tokenizers.Tokenizer\`) backend has \`special=False\`: Output (verified on CI runner): ## Why the discrepancy exists The...huggingface/transformersydshiehSep 92 weeks ago
Kernels for partial RoPE### Feature request Would be cool to add kernel for partial rotations like below ### Motivation @vasqu 😄 (https://github.com/huggingface/transformers/pull/48594#discussion_r3959426411) ### Your contribution tell me if you need anythinghuggingface/transformerszucchini-nlpSep 82 weeks ago
Loading lugh/ClinicalT5-baseI load my model the following way:. from transformers import AutoTokenizer, AutoModelForSeq2SeqLM model_id = "luqh/ClinicalT5-base" tokenizer = AutoTokenizer.from_pretrained(model_id) model = AutoModelForSeq2SeqLM.from_pretrained(model_id, from_flax=True) However I get the...Stack Overflow [huggingface-transformers]ace-of-spadesSep 23 weeks ago
[Bug] `HfArgumentParser` silently corrupts `tuple` and `set` typed dataclass fields from CLI...with `tuple` or `set` types (e.g. `tuple[int, int]` or `set[str]`), `HfArgumentParser` fails to parse them correctly, resulting in silent data corruption or unparsed argument errors: ### Root Cause Analysis In `src/transformers/hf_argparser.py` (lines 220-234),...huggingface/transformersGracy769Aug 303 weeks ago
Nested per-layer `rope_parameters` is misparsed when `layer_types` omits one of its keys### System Info - `transformers`: `main` @ `83d46aa2a2` (reported version `5.16.0.dev0`). The relevant code in `modeling_rope_utils.py` is identical at the `v5.16.1` tag, so the released version is affected too. - `torch`: 2.8.0+cu128 - Python: 3.12.9 - Platform:...huggingface/transformersalbertvillanovaAug 284 weeks ago
Vectorized (batched) logits processors for assisted decoding### Feature request Vectorize the logits processor application over draft slots in assisted decoding, so speculative candidate generation can apply processors to all drafted positions in one call instead of looping one draft step at a time. ### Context This came up during review...huggingface/transformersVaggelisGianAug 284 weeks ago

Companies and products named

Company or productPosts naming it
Hugging Face15
GitHub4
PyTorch1
NVIDIA1
About this problem

Evidence

15 posts from 15 people in 3 places, about 3 a week over 29 days. Mostly on huggingface/transformers, Hugging Face Forums, Stack Overflow [huggingface-transformers]. Tools named alongside: GitHub, PyTorch, NVIDIA.

Frustration Frustration 0 of 3· Seen on GitHub issues, Community forums, Stack Overflow

How it was grouped

Posts about Hugging Face that state a pain but match no known issue yet. First post Aug 28, 2026, latest Sep 25, 2026. Corroborated: 3 or more posts from 2 or more people or places. Method

History

  • 2026-09-25 Statement: Hugging Face users keep asking for help to Hugging Face users keep filing bugs
  • 2026-09-25 Added: Hugging Face users keep asking for help (15 posts)

Rising problems by email

Mondays: the problems in data, tech and AI that grew fastest that week.

Double opt-in. Unsubscribe any time.