All issues

Search results

Add RWKV-7 (Goose)

Status: Open.

#47787 In huggingface/transformers;

· Hakureirm opened on Aug 5, 2026

[serge] integration failure triage - 2026-08-04

Status: Open.

#47783 In huggingface/transformers;

· github-actions opened on Aug 4, 2026

incorrect precedence of generation_config values

bug

Status: Open.

#47752 In huggingface/transformers;

· jkl-ds opened on Aug 3, 2026

HuBERT positional convolution is sensitive to padding when conv_pos_batch_norm=True

Status: Open.

#47739 In huggingface/transformers;

· Abdulrhman-shaheen opened on Aug 3, 2026

Add native support for OpenBMB MiniCPM4

New model

Status: Open.

#47732 In huggingface/transformers;

· sushmanthreddy opened on Aug 3, 2026

[Model Request] Native support for THUDM CogVLM2-Video

New model

Status: Open.

#47713 In huggingface/transformers;

· AryanSharma21 opened on Aug 2, 2026

RobertaTokenizerFast.from_pretrained() loads broken tokenizer in v5 (missing pre_tokenizer and BPE merges)

Status: Open.

#47706 In huggingface/transformers;

· alwaysprince05 opened on Aug 2, 2026

Gemma-4-E4B: the PLE table holds ~5.6 GB of VRAM that can live on CPU bit-identically - is there a supported device_map path?

Status: Open.

#47705 In huggingface/transformers;

· MyceliaLabsBV opened on Aug 2, 2026

Add native support for OpenBMB VoxCPM2

New model

Status: Open.

#47695 In huggingface/transformers;

· sushmanthreddy opened on Jul 31, 2026

Add native support for Amazon Chronos-2

New model

Status: Open.

#47692 In huggingface/transformers;

· sushmanthreddy opened on Jul 31, 2026

Heads that declare **kwargs but drop num_items_in_batch train with an inflated loss under gradient accumulation

Status: Open.

#47688 In huggingface/transformers;

· Yigtwxx opened on Jul 31, 2026

Grounding DINO text position embeddings are truncated to int64, collapsing 93% of them to zero

Status: Open.

#47674 In huggingface/transformers;

· Yigtwxx opened on Jul 31, 2026