Search results

[Model Request] Add EUPE (Efficient Universal Perception Encoder) by Meta AI

New model

Status: Open.

#45322 In huggingface/transformers;

· forensicmike opened on Apr 8, 2026

Qwen3.5: DeepSpeed ZeRO-3 fails to load weights for language_model

Status: Open.

#45313 In huggingface/transformers;

· debOliveira opened on Apr 8, 2026

[BUG] transformers>=5.4.0, Qwen3.5 Moe from_pretrained error

bug

Status: Open.

#45310 In huggingface/transformers;

· Jintao-Huang opened on Apr 8, 2026

Feature request: Support evaluation every N epochs in TrainingArguments

Feature requestRequest for a new featureRequest for a new feature

Status: Open.

#45308 In huggingface/transformers;

· varuna-km-18267 opened on Apr 8, 2026

Bug, Generation

Status: Open.

#45307 In huggingface/transformers;

· Regata3010 opened on Apr 8, 2026

Remove import * usage

Status: Open.

#45306 In huggingface/transformers;

· yueyinqiu opened on Apr 8, 2026

Gradients not averaged by GAS when using DeepSpeed + model_accepts_loss_kwargs=True (Qwen3, Llama3, etc.)

bug

Status: Open.

#45305 In huggingface/transformers;

· florian6973 opened on Apr 7, 2026

NexusQuant: training-free KV cache compression (10-33x) via DynamicCache hooks

Status: Open.

#45304 In huggingface/transformers;

· jagmarques opened on Apr 7, 2026

Support Sequence Classification for Gemma 4 Models

Feature requestRequest for a new featureRequest for a new feature

Status: Open.

#45295 In huggingface/transformers;

· jesperschlegel opened on Apr 7, 2026

resize_token_embeddings does not effect to output_embeddings

bug

Status: Open.

#45292 In huggingface/transformers;

· KoichiYasuoka opened on Apr 7, 2026

apply_chat_template(tokenize=True) crashes on assistant messages with tool calls and no content

bug

Status: Open.

#45290 In huggingface/transformers;

· qgallouedec opened on Apr 7, 2026

Many import errors after update from 4.57.0 to 5.5.0

bug

Status: Open.

#45278 In huggingface/transformers;

· marchcat69 opened on Apr 7, 2026