Issues

Revalidate Transformer Engine GPU CI support

huggingface/peft#3983 · opened

on Oct 14, 2026

OFT Conv2d: unmerged and merged outputs differ for kernels larger than 1x1

huggingface/peft#3982 · opened

on Oct 14, 2026

Adding or loading a Trainable Tokens adapter while another one is merged corrupts the base embeddings

huggingface/peft#3979 · opened

on Oct 14, 2026

Native TransformerEncoder fast path ignores trained PEFT adapters during evaluation

huggingface/peft#3976 · opened

on Oct 14, 2026

OFT on nn.Linear fails for 4-D inputs (e.g. ConvNeXt pwconv layers)

huggingface/peft#3968 · opened

on Oct 14, 2026

1

Mixed adapter batch raises KeyError when only one adapter has modules_to_save

huggingface/peft#3967 · opened

on Oct 14, 2026

An old, unused torchao install makes every LoRA/OFT injection fail with ImportError (e.g. on Colab and Kaggle)

huggingface/peft#3964 · opened

on Oct 13, 2026

2

LoRA-FA optimizer.step(closure) runs backward with gradients disabled

huggingface/peft#3958 · opened

on Oct 13, 2026

1

`unload()` returns the trained `modules_to_save` copy instead of the original module

huggingface/peft#3955 · opened

on Oct 13, 2026

1

Prompt-learning methods crash on transformers 4.43/4.44: the pre-get_text_config fallback uses `in` on a PretrainedConfig

huggingface/peft#3951 · opened

on Oct 13, 2026

MoE conversion no longer targets routed experts when shared experts or dense layers share the target name

huggingface/peft#3945 · opened

on Oct 12, 2026

1

LoRA on nn.MultiheadAttention: lora_bias=True never trains the bias, and forward fails on a bias-free MHA

huggingface/peft#3935 · opened

on Oct 11, 2026

2

get_peft_model_state_dict gathers the whole sharded base model on DTensor-sharded models

huggingface/peft#3930 · opened

on Oct 11, 2026

5

`ensure_weight_tying=True` silently stops tying the LoRA weights when the model is in bf16/fp16

huggingface/peft#3929 · opened

on Oct 11, 2026

CPT forward mutates the caller-provided input_type_mask in place

huggingface/peft#3928 · opened

on Oct 11, 2026

1

`merge_and_unload()`/`unload()` raise AttributeError on RoBERTa sequence classifiers for all methods except LoRA, AdaLoRA and HiRA

huggingface/peft#3918 · opened

on Oct 9, 2026

1

LoRA on MoE experts (target_parameters) fails under tensor/expert parallelism

huggingface/peft#3914 · opened

on Oct 9, 2026

2

LoRA-FA optimizer crashes when an adapted layer has no gradient

huggingface/peft#3900 · opened

on Oct 6, 2026

1

Default CPTConfig masks all input labels and returns NaN training loss

huggingface/peft#3899 · opened

on Oct 6, 2026

3

Inference-only prompt adapters remain trainable after loading

huggingface/peft#3898 · opened

on Oct 6, 2026

1

Adapters are saved under transformers v5's renamed module paths, while `save_pretrained` writes base models in the original checkpoint layout

huggingface/peft#3896 · opened

on Oct 5, 2026

9

CPT loss uses the default adapter config instead of the active adapter config

huggingface/peft#3879 · opened

on Oct 4, 2026

4

Add support for BoRA

huggingface/peft#3877 · opened

on Oct 3, 2026

1

`get_peft_model_state_dict` returns an empty dict for FSDP1-wrapped models since 0.21.0 (adapter checkpoints / weight syncs silently empty)

huggingface/peft#3876 · opened

on Oct 3, 2026

4

MoE recipe in the LoRA docs gives adapters that vLLM serves with the wrong expert scaling

huggingface/peft#3874 · opened

on Oct 2, 2026

1

Navigation Menu

Repository navigation

Issues

Revalidate Transformer Engine GPU CI support

OFT Conv2d: unmerged and merged outputs differ for kernels larger than 1x1

Adding or loading a Trainable Tokens adapter while another one is merged corrupts the base embeddings

Native TransformerEncoder fast path ignores trained PEFT adapters during evaluation

OFT on nn.Linear fails for 4-D inputs (e.g. ConvNeXt pwconv layers)

Mixed adapter batch raises KeyError when only one adapter has modules_to_save

An old, unused torchao install makes every LoRA/OFT injection fail with ImportError (e.g. on Colab and Kaggle)

LoRA-FA optimizer.step(closure) runs backward with gradients disabled

`unload()` returns the trained `modules_to_save` copy instead of the original module

Prompt-learning methods crash on transformers 4.43/4.44: the pre-get_text_config fallback uses `in` on a PretrainedConfig

MoE conversion no longer targets routed experts when shared experts or dense layers share the target name

LoRA on nn.MultiheadAttention: lora_bias=True never trains the bias, and forward fails on a bias-free MHA

get_peft_model_state_dict gathers the whole sharded base model on DTensor-sharded models

`ensure_weight_tying=True` silently stops tying the LoRA weights when the model is in bf16/fp16

CPT forward mutates the caller-provided input_type_mask in place

`merge_and_unload()`/`unload()` raise AttributeError on RoBERTa sequence classifiers for all methods except LoRA, AdaLoRA and HiRA

LoRA on MoE experts (target_parameters) fails under tensor/expert parallelism

LoRA-FA optimizer crashes when an adapted layer has no gradient

Default CPTConfig masks all input labels and returns NaN training loss

Inference-only prompt adapters remain trainable after loading

Adapters are saved under transformers v5's renamed module paths, while `save_pretrained` writes base models in the original checkpoint layout

CPT loss uses the default adapter config instead of the active adapter config

Add support for BoRA

`get_peft_model_state_dict` returns an empty dict for FSDP1-wrapped models since 0.21.0 (adapter checkpoints / weight syncs silently empty)

MoE recipe in the LoRA docs gives adapters that vLLM serves with the wrong expert scaling