-
Notifications
You must be signed in to change notification settings - Fork 772
Open
Description
Describe the feature
Update the list of supported liger kernels to include patches for newer models (Qwen3
, Gemma-3
, ...)
Paste any useful information
The list of liger_kernels
supported by ms-swift
ms-swift/swift/llm/train/tuner.py
Lines 22 to 26 in a26c6a1
from liger_kernel.transformers import (apply_liger_kernel_to_llama, apply_liger_kernel_to_mistral, | |
apply_liger_kernel_to_mixtral, apply_liger_kernel_to_gemma, | |
apply_liger_kernel_to_qwen2, apply_liger_kernel_to_qwen3, | |
apply_liger_kernel_to_qwen2_vl, apply_liger_kernel_to_qwen2_5_vl, | |
apply_liger_kernel_to_phi3, apply_liger_kernel_to_mllama) |
Missing ones include Qwen3: liger_kernel.transformers.apply_liger_kernel_to_qwen3
and Gemma-3: liger_kernel.transformers.apply_liger_kernel_to_gemma3
.
Metadata
Metadata
Assignees
Labels
No labels