[Inquiry]: Unable to use PEFT for training

[moyitech]

Basic Information - Models Used

MiniMax-M2

Description

When I try to load it using bf16, the printed model still shows the linear layers as fp8, and it cannot be dequantized.

If you don't dequantize, the Peft process will display the error message: "The model you are trying to fine-tune is quantized with QuantizationMethod.FP8, but that quantization method does not support training."

[AkaTsukijm]

same:
ValueError: The model you are trying to fine-tune is quantized with fp8 but that quantization method do not support training. Please open an issue on GitHub: https://github.com/huggingface/transformers to request the support for training support for fp8'

it is because they dont want us to finetune on their model?